Treatment of Shigellosis: III. Comparison of One- or Two-Dose Ciprofloxacin with Standard 5-Day Therapy A Randomized, Blinded Trial Michael L. Bennish, MD; Mohammed Abdus Salam, MB, BS; Wasif Ali Khan, MB, BS; Ali Miraj Khan, MB, BS
• Objective: To determine whether a single dose, or 2 doses, of ciprofloxacin are as effective as 5-day, 10dose therapy for the treatment of shigellosis in adult men who are moderately to severely ill. • Design: Randomized, double-blind clinical trial. • Setting: A diarrhea treatment center in the capital city of a developing country, Bangladesh. • Patients: A total of 128 adult men with dysentery of less than 96 hours duration. All had Shigella organisms isolated from a culture of stool. • Interventions: Patients were randomly assigned to receive either a single 1-gram dose of ciprofloxacin at admission to the study (single-dose group; n = 40), a 1 -gram dose of ciprofloxacin at admission and 24 hours later (2-dose group; n - 43), or 500 mg of ciprofloxacin every 12 hours for 5 days (10 dose group; n = 35). All patients were hospitalized for 6 days. • Measurements: Stools were collected individually; their character and consistency were recorded and cultured daily. A physical examination and recording of symptoms were done daily, and the temperature was measured every 4 hours. Therapy was considered to have failed in patients who did not have improvement in the signs and symptoms of dysentery after 72 hours of therapy or in patients who on study day 5 had more than nine stools, or more than two watery stools, or were febrile. • Results: There were no treatment failures in the 78 patients infected with species of Shigella other than Shigella dysenteriae type 1. Among the 40 patients infected with S. dysenteriae type 1, treatment failed in 4 of the 10 patients who received single-dose therapy, 2 of the 15 patients who received 2-dose therapy, and none of the 15 patients who received 10-dose therapy ( P = 0.017, single-dose therapy group compared with 10-dose group; P = 0.15 for the single-dose group compared with the 2-dose group; P> 0.2 for the 2-dose group compared with the 10-dose group). • Conclusions: A single 1-gram dose of ciprofloxacin is effective therapy for patients infected with species of Shigella other than S. dysenteriae type 1. Single-dose therapy is inferior to 10-dose therapy for treating patients infected with S. dysenteriae type 1.
Annals of Internal Medicine. 1992;117:727-734. From the International Centre for Diarrhoeal Disease Research, Bangladesh, Dhaka, Bangladesh; and New England Medical Center, Boston, Massachusetts. For current author addresses, see end of text.
^Shigellosis remains a major cause of diarrhea-related morbidity and mortality in developing countries (1-3). Antimicrobial therapy is indicated for all persons infected with Shigella spp. who have symptoms of dysentery (4), and effective treatment of such patients reduces resultant morbidity and mortality (4, 5). A number of factors limit the effective therapy of shigellosis. First, in many countries, Shigella strains are frequently resistant to ampicillin, trimethoprim-sulfamethoxazole, and nalidixic acid (6-12), the drugs of choice for treating this infection (4, 13). Resistance is most common among strains of Shigella dysenteriae type 1, which is endemic or epidemic in South and Southeast Asia, Central Africa, and Central America (1, 2, 6, 8, 9, 11, 14). Second, in developing countries, where severe shigellosis is most common, drugs are most often dispensed by persons with minimal—or no—formal medical training (15-17). A full course of therapy is often not prescribed for infections; most commonly, medical practitioners dispense, or prescribe, only a few doses of medication (15, 16). Third, the cost of therapy, especially if newer agents are required, can be prohibitively expensive for persons in developing countries, where the per capita income is often less than $1 per day (18). A short course of ciprofloxacin therapy, if effective, would obviate a number of these problems. First, most multiresistant strains of Shigella remain susceptible to the newer quinolones, including ciprofloxacin (19-21). Second, if a single dose—or two doses—of the drug were effective in treating shigellosis, the failure of most dispensers to provide a 5-day course of therapy would be less of an obstacle to providing effective therapy. Third, patient compliance is likely to be greater than with a conventional 10-dose regimen. Finally, the cost of treatment could be reduced. A short course of ciprofloxacin therapy might be effective in treating shigellosis for several reasons. The newer quinolones, including ciprofloxacin, are known to
SI Unit Conversion Factor Normal Range Blood Leukocyte Count Traditional: 3200 - 9800 mm - 3 mm ~3 x 0.001 SI: 3.2 - 9.8 x 109/L = 109/L Drug Brand Name Generic Name Cipro ciprofloxacin
© 1992 American College of Physicians
Downloaded From: http://annals.org/pdfaccess.ashx?url=/data/journals/aim/19762/ by a University of California San Diego User on 06/03/2017
727
be effective in treating shigellosis when given for 3 to 5 days (14, 19, 22, 23). In relation to the concentration of the drug required to inhibit growth of Shigellae, ciprofloxacin attains high concentrations in serum (mean peak concentrations more than 600 times the minimum inhibitory concentration [MIC] in following a 500-mg dose) and in stool (concentrations uniformly more than 1000 times the MIC) (19). The serum half-life of ciprofloxacin is relatively long, averaging 3 to 5 hours (24, 25). Finally, a single dose of 800 mg of norfloxacin has been found to be effective in treating adults with diarrhea or dysentery of mild severity caused by infection with Shigella flexneri or Shigella sonnei (26). In this study we compared the effectiveness of a single dose of 1 gram of ciprofloxacin, two doses of 1 gram of ciprofloxacin given on successive days, and 500 mg of ciprofloxacin given twice daily for 5 days, on the resolution of symptoms in adult men moderately to severely ill with shigellosis due predominantly to infections with either 5. dysenteriae type 1 or S. flexneri. Methods Patient Population and Selection of Study Patients This study was conducted at the Diarrhoea Treatment Centre of the International Centre for Diarrhoeal Disease Research, Bangladesh, from August 1989 to December, 1990. Patients eligible for study had to fulfill the following criteria:
Table 1. Characteristics of tbte Three 1rreatmentt Groups at Admission* Characteristic Patients, n Age,yt Duration of illness, h Stool frequency before therapy on day of admission^ Vomited before admission Weight, Ag§ Temperature > 37.8° C, admission day, n (%) Mild or moderate dehydration, n(%) Abdominal tenderness, n (%) Shigella sp. isolated, n (%) S. flexneri S. dystenteriae type 1 Other species Other enteric pathogens, n (%) Salmonella sp. Plesiomonas shigelloides or Aeromonas sp. Campylobacter jejuni C. jejuni and Plesiomonas shigelloides Stool microscopic values, n(%) > 50 leukocytes per high power field > 50 erythrocytes per high power field
Treiatment Gi^HE 1-Dose 2-Dose 10-Dose 40 43 35 31 ± 10 26 ± 7 32 ± 10 50 ± 2 0 46 ± 2 1 52 ± 2 0 17 ± 14 21 ± 18 22 ± 16 12 (30) 12 (28) 11 (31) 47 ± 5 50 ± 7 50 ± 9 16 (40)
18 (42)
7(20)
13 (32) 35 (87)
11 (26) 34 (79)
13 (37) 32 (91)
22 (55) 19 (44) 10 (25) 15 (35) 8(20) 9(21)
18 (51) 15 (43) 2(6)
1(3)
1(2)
1(3)
2(3) 0
6(14) 2(5)
2(6) 0
0
0
1(3)
37 (93)
41 (95)
34 (97)
36 (90)
38 (88)
34 (97)
* Values are mean (± SD) unless noted. t P ~ 0.008. One-dose group comf>ared with 2-•dose group, P = 0.02; 2-dose group compared with 10-dose\ group, P =• 0.004. % Determined by history. § P = 0.06. One-dose group compared with 2-dose group., P = 0.04; 1-dose group compared with 10-dosek. group, P »= 0.04. 728
signs and symptoms of dysentery, as manifested by the presence of gross blood and mucus on visual inspection of a stool sample and a history of tenesmus and abdominal pain; male gender; age, 18 to 60 years; illness duration less than 96 hours; no acute illness other than dysentery; and absence of trophozoites of Entamoeba histolytica on microscopic examination of stool. Women were excluded from the study because of concerns about the ability to detect those who were early in a pregnancy and because of social constraints limiting a woman's ability to remain hospitalized for the full 6-day study period. Study Protocol At admission, a complete history and physical examination were done and the following laboratory tests obtained: culture of stool and rectal swab samples for Shigella spp., Salmonella spp., Vibrio spp., Plesiomonas shigelloides, Aeromonas sp., and Campylobacter jejuni; microscopic examination of stool for enumeration of leukocytes and erythrocytes and identification of cysts and ova of parasites; and a complete blood count. All patients were enrolled in the morning and received their first dose of medication at 1400 hours. A study day was considered to begin at 1400 hours and to continue until 1400 hours the next day. Patients were randomly assigned to one of three treatment groups: ciprofloxacin (Cipro; Bayer, Leverkusen, Germany), 500 mg every 12 hours for 5 days (10-dose group); 1 gram of ciprofloxacin given at admission (single-dose group); and 1 gram of ciprofloxacin given at admission and a repeat dose of 1 gram given 24 hours later (two-dose group). Neither the patients nor the staff were aware of which therapy was being given. Medication was provided in blister packages that contained either active drug or placebo, which were identical in appearance and similar in taste. Medication was assigned to a study number using a random number table and a block randomization method with a block size of nine. Patients were hospitalized for 6 full days after the first dose of medication. Symptoms and physical examination findings were recorded daily. Vital signs, including oral temperature, were measured every 4 hours. Fever was defined as an oral temperature greater than 37.8 °C. Stools were collected separately on plastic-lined pads and examined for consistency and the presence of blood or mucus. Stool consistency was graded as watery, soft, or formed according to previously defined criteria (19). Individual stools of varied consistency were categorized according to their predominant characteristic. For each study day, the consistency and character of stools was categorized according to the worst finding in any individual stool sample for that day. A rectal swab or stool sample for culture was obtained daily. Microscopic examination of stool was repeated on study days 3 and 5. Patients were requested to return for evaluation 7 days after discharge. At the return visit a history and physical examination were done, and a stool or rectal swab sample was obtained for culture. Culture and microscopic examination of stool, susceptibility testing, and blood counts were done according to previously described methods (19, 27). Evaluation of Outcome Afebrile patients were considered to have resolution of illness if, on the last day of therapy (study day 5), they had 3 or fewer stools, none watery; to have marked improvement if they had 6 or fewer stools, 1 or fewer watery; to have slight improvement if they had 9 or fewer stools, 2 or fewer watery. Treatment was considered to have failed if, on study day 5, patients were febrile, or had 10 or more stools, or had 3 or more watery stools. Therapy was also considered to have failed in patients who at the end of study day 3 continued to have only frankly bloody-mucoid stools, persistent fever, or severe abdominal pain. All patients in whom therapy failed were treated with ciprofloxacin, 500 mg every 12 hours, for 5 days. Treatment was provided to these patients without knowing which treatment they had previously received. Patients were considered bacteriologically cured if the infecting strain
1 November 1992 • Annals of Internal Medicine • Volume 117 • Number 9
Downloaded From: http://annals.org/pdfaccess.ashx?url=/data/journals/aim/19762/ by a University of California San Diego User on 06/03/2017
of Shigella could not be cultured from a stool or rectal swab sample on study day 3 or thereafter.
Table 2. Clinical Ou tcome by Treatmc;nt Group and Species of Shigella*
Statistical Methods
Shigella Species
Data entry and significance testing were done using a database and statistical program for personal computers (Stat-Pac Gold, version 3.2, Walonick Associates, Minneapolis, Minnesota). Analysis of variance was used to evaluate the significance of differences between the three study groups for continuous variables that were normally distributed; if the difference was significant at P < 0.05, then individual groups were compared using the Students f-test. Differences in continuous variables that were not randomly distributed, such as duration of symptoms, were tested for significance using the Kruskal-Wallis test, and individual groups then were compared using the Mann-Whitney U test if P < 0.05 on the KruskalWallis test. All tests were two-tailed. The significance of differences in proportions was tested using a 2 x 3 chi-square test. If P < 0.05 on the 2 x 3 chi-square test, then individual groups were compared using a 2 x 2 chi-square test, or Fisher exact test if the predicted size of any cell was 5 or less. The initial analysis of the data was blinded; the treatment groups were labeled as 1, 2, and 3 without knowledge of which treatment each group had received. Written informed consent was obtained from all patients. The study protocol was approved by the Ethical and Research Review Committees of the International Centre for Diarrhoeal Disease Research, Bangladesh. Results Patient Enrollment and Admission Characteristics One hundred sixty-two patients were enrolled in the study, 54 in each of the three treatment groups. Thirtyfour (21.0%) patients (10 in the 1-dose group, 8 in the 2-dose group, and 16 in the 10-dose group; P = 0.14) did not have Shigella organisms isolated from an admission rectal swab or stool sample. Ten (7.8%) of the 128 culture-positive patients were excluded from analysis, nine because they withdrew from the study before completion of therapy and one patient because he developed symptomatic rotavirus infection on study day 5. Patients who withdrew from the study did so because they decided not to remain in the hospital for the full study period. Of the nine patients who withdrew from the study, seven did so on the first study day. Four of the 10 excluded patients were in the single-dose group, three in the 2-dose group, and 3 in the 10-dose group (P = 0.2). Thus, of the 154 patients initially entered into the study, 118 patients were eligible for primary analysis comparing the efficacy of the three treatment regimens in patients with culture-confirmed shigellosis. Forty (34%) of these 118 patients were in the singledose treatment group; 43 (36%) in the 2-dose group; and 35 (30%) in the 10-dose group. The admission clinical characteristics of these 118 patients are shown in Table 1. Shigella flexneri (50% of patients) and S. dysenteriae type 1 (34% of patients) were the most common Shigella species isolated from patients. The mean age of patients in the 2-dose treatment group was 26 years, significantly (P < 0.05) less than the mean age of patients in the single-dose and 10-dose groups (mean ages 31 and 32 years, respectively). The three treatment groups did not differ significantly in any other admission characteristic. No patient had clinical signs of malnutrition. Six patients (three in the single-dose group, one in the
Outcoime Resolution Marked Slight Failure Improve- Improvement ment
5(50) 6(40) 5(33)
1(10) 4(27) 8(53)
0 3 (20) 2 (13)
4 (40) 2 (13) 0
24 (80) 25 (89) 18 (90)
6(20) 3(H) 2(10)
0 0 0
0 0 0
29 (72) 31 (72) 23 (66)
7(18) 7(16) 10 (29)
0 3 (7) 2(6)
4 (10) 2 (5) 0
* Clinical outcome was determined oil study day 5. t For S. dysenteriae type$ 1-infected \jatients, 3 )< 4 chi-square test, P = 0.05. Proportion of \>atients in vvhom treatinent failed, 1-dose compared with 10-dose groilp, P = 0.01 7, Fisher ex:act test. Proportion of patients in whom treatnlent failed, 1 -dose and \2-dose groups compared with 10-dose group, iP = 0.05, Fi sher exact itest. % For patients infected with species of Shigella o ther than S. dysenteriae type 1, 3 x 4 chi-sqiiiare test, P > 0.2. § For all patients, 3 x 4 chi-square t 0.:
2-dose group, and two in the 10-dose group), infected with a nalidixic-acid-susceptible strain of Shigella, had received nalidixic acid before study entry; none had received more than four capsules. One patient in the 2-dose group was infected with an ampicillin-susceptible strain of Shigella and had taken four capsules of ampicillin before admission. Six of the seven patients who had previous therapy were infected with S. flexneri, compared with 53 (48%) of the 111 patients who had not received therapy (P = 0.06), and patients who had previous therapy had higher blood leukocyte counts at admission than did patients who had not received therapy (15.2 x 109/L compared with 11.2 x 10 9 /L, P = 0.002). The two groups had no other significant differences in pre-admission characteristics. Excluding the seven patients who had received therapy before entry into the study did not affect the differences among the three treatment groups in any of the major outcome variables.
Clinical Outcome Treatment failed in six patients, four who had received single-dose therapy and 2 who had received 2-dose therapy (Table 2) (P = 0.12, single-dose compared with 10-dose group; P > 0.2, 2-dose compared with 10-dose group). Four of these six patients received open therapy before the end of study treatment because of continued dysenteric symptoms. The remaining two patients in whom treatment failed each had more than 10 stools on the last day of therapy and received open therapy beginning on study day 6. All six patients in whom therapy failed were infected with S. dysenteriae tvne 1. comnared with 34 (30.4%^ of the 112 Datients in
1 November 1992 • Annals of Internal Medicine • Volume 117 • Number 9
Downloaded From: http://annals.org/pdfaccess.ashx?url=/data/journals/aim/19762/ by a University of California San Diego User on 06/03/2017
729
ure 3). Bacteriologic failure occurred in a single patient, who was in the 2-dose therapy group. Shigella dysenteriae type 1 was isolated from a stool sample from this patient on the first three study days, and he was started on open ciprofloxacin therapy on study day 4 because of lack of clinical response. Outcome in Shigella dysenteriae Type 1-Infected Patients
Figure 1. Mean stool frequency by treatment group and study day. The error bars indicate the upper 95% CI of the mean. No statistically significant differences between study groups in the mean number of stools were found for any of the study days. The mean number of stools before admission and on follow-up were determined by history. The pre-admission period is the period from the midnight before admission until admission to the study. For follow-up, the time period was the 24 hours before the visit to the treatment center. For study days 1 to 6, n = 40 in the single-dose group; 43, in the 2-dose group; and 35, in the 10-dose group. At follow-up, n = 34 in the singledose group; 41, in the 2-dose group; and 34, in the 10-dose group. Four patients in the 1-dose group and two patients in each of the 2- and 10-dose groups received open-label, 5-day therapy with ciprofloxacin starting on study day 4 or later. Adm = admission to study; F-up = follow-up.
whom therapy was successful (P = 0.001). Patients in whom therapy failed also had a history of statistically significantly more stools on the day of, and on the day before, admission to the study and had a higher systolic blood pressure when admitted. During the 6 study days, patients in the single-dose group had a mean of 34 stools: those in the 2-dose group, 37 stools; and those in the 10-dose group, 36 stools (P = 0.09) (Figure 1). Only one patient, who was in the single-dose group, was still febrile after 48 hours of therapy (Figure 2). Differences among the study groups in the proportion of patients with abdominal tenderness or tenesmus were not significant for any of the study days. Mucus was present in the stool of patients in the 10-dose group for a longer period (median 6 days) than in patients in the single-dose treatment group (median, four days, P = 0.014) (Table 3). The median duration of other stool characteristics was similar for the three treatment groups. The number of leukocytes and erythrocytes in the stool of the patients in the three treatment groups did not differ significantly on the 3 days (admission, day 3, and day 5) that the test was done.
Bacteriologic Response Shigella organisms were more rapidly eradicated from stool cultures in patients receiving short-course therapy than in those receiving conventional 5-day therapy (Fig730
The 40 patients with S. dysenteriae type 1 disease had more severe illness than did the 78 study patients infected with other species of Shigella. Shigella dysenteriae type 1 infected-patients had significantly (P < 0.05) more stools on each of the 3 days before initiation of the study, and on each of the 6 study days. These patients had a mean of 52 stools during the study, compared with 27 in patients infected with other Shigella species (P < 0.001). Bacterial resolution was also slower in S. dysenteriae type 1 infected-patients; Shigella was isolated in 12 (30%) of 40 5. dysenteriae type 1-infected patients after 1 day of therapy compared with 9 (12%) of 78 patients infected with other Shigella spp. (P = 0.03); after study day 2, the respective figures were seven (18%) compared with two (3%, P = 0.007). All six patients in whom treatment failed and the five patients who had only slight improvement after completion of therapy were among the 40 study patients infected with S. dysenteriae type 1 (see Table 2). Treatment failed in a larger proportion of patients infected with S. dysenteriae type 1 who received single-dose therapy (4 of 10) than in patients who received 10-dose therapy (0 of 15 patients, P = 0.017). The rate of treatment failure in the 2-dose group, 2 of 15 patients, did
Figure 2. Number of patients with fever by treatment group and study day. Fever was defined as an oral temperature greater than 37.8 °C. Differences between the groups were not significant for any of the study days. For study day 1 to 6, n = 4 0 in the single-dose group; 43, in the 2-dose group; and 35, in the 10-dose group. At follow-up, n = 34 in the single-dose group; 41, in the 2-dose group; and 34, in the 10-dose group. Four patients in the 1-dose group and two patients in each of the 2and 10-dose groups received open-label, 5-day therapy with ciprofloxacin starting on study day 4 or later. Adm = admission to study; F-up = follow-up.
1 November 1992 • Annals of Internal Medicine • Volume 117 • Number 9
Downloaded From: http://annals.org/pdfaccess.ashx?url=/data/journals/aim/19762/ by a University of California San Diego User on 06/03/2017
Table 3. Duration c>f Findings by Treatment Group Finding
Treatment Group 1 Dose 2 Dose 10 Dose (n = 40) (n = 43) (n = 35) \jt
0.2). The combined failure rate in the two short-course treatment groups (6 of 25 patients) was significantly (P = 0.05) higher than for patients treated with 10 doses. Shigella dysenteriae type 1-infected patients who failed therapy or who had only slight improvement in symptoms did not differ from other S. dysenteriae type 1-infected patients in duration of illness or stool frequency before therapy. Early in the course of illness, S. dysenteriae type 1-infected patients treated with a single dose of ciprofloxacin had fewer symptoms than did patients treated with conventional 10-dose therapy. After the first day of therapy, patients in the single-dose group less commonly had abdominal pain (6 of 10 patients compared with 15 of 15 patients, P = 0.016), joint pain (2 of 10 patients compared with 10 of 15 patients, P = 0.03), and headache (1 of 10 patients compared with 10 of 15 patients, P = 0.007) than did patients in the 10-dose group. Patients who received 2-dose therapy less frequently had headache or joint pain on the first study day than did the conventional therapy group. On study day 2, anorexia (4 of 10 patients compared with 13 of 15 patients, P = 0.02) and abdominal pain (5 of 10 patients compared with 15 of 15 patients, P = 0.005) were less common in the single-dose group than in the 10-dose group.
of the 78 non-5, dysenteriae type 1 isolates were resistant to ampicillin, 22 (28%) were resistant to trimethoprim-sulfamethoxazole, and only 1 (1%) isolate was resistant to nalidixic acid (P < 0.001 for all three comparisons to S. dysenteriae type 1 isolates). All isolates were susceptible to amdinocillin pivoxil (mecillinam).
Follow-up One-hundred nine of the 118 patients returned for a follow-up evaluation a mean of seven days after discharge from the treatment center. One patient in the single-dose treatment group, who had S. flexneri isolated at admission to the study, but whose stool culture was negative on discharge, had S. flexneri again isolated at follow-up; one other patient in the single-dose group, who had been infected with 5. flexneri at admission to the study, had S. sonnei isolated on his return visit. An additional patient in the single-dose group had recurrent symptoms of dysentery at follow-up; C. jejuni, which was not isolated from an admission or day 5 stool culture, was isolated from a stool culture at follow-up.
Outcome in Patients Not Infected with Shigella Thirty-four patients initially enrolled in the study did not have Shigella isolated from a stool or rectal swab culture. Seven of these patients were infected V. cholera (one with the 0 1 serogroup and six with other serogroups); five with Aeromonas sp., and two each with C. jejuni and P. shigelloides. Patients were dis-
Susceptibility of the Shigella Isolates All Shigella isolates were susceptible to ciprofloxacin when tested by both the disc-diffusion and broth dilution methods. The median MIC of ciprofloxacin for the 35 S. dysenteriae type 1 isolates tested was 0.060 /ig/ mL; for the 59 isolates of other Shigella species tested, it was < 0.004 /ig/mL (P < 0.001). All 40 isolates of S. dysenteriae type 1 were resistant to both ampicillin and trimethoprim-sulfamethoxazole when tested by the disc-diffusion method; 31 (78%) of the 40 isolates were also resistant to nalidixic acid. In comparison, 31 (40%)
Figure 3. Isolation rates of Shigella from stool or rectal swab samples by treatment group and study day. The asterisk indicates that for study day 2, P = 0.04, 2 x 3 chi-square test. P = 0.04, 1-dose compared with the 10-dose group. P = 0.076, 2-dose compared with 10-dose group. For study days 1 to 6, n = 40 in the single-dose group; 43, in the 2-dose group; and 35, in the 10-dose group. At follow-up, n = 34 in the single-dose group; 41, in the 2-dose group; and 34, in the 10-dose group. Four patients in the 1-dose group and two patients in each of the 2- and 10-dose groups received openlabel, 5-day therapy with ciprofloxacin starting on study day 4 or later. Adm = admission to study; F-up = follow up.
1 November 1992 • Annals of Internal Medicine • Volume 117 • Number 9
Downloaded From: http://annals.org/pdfaccess.ashx?url=/data/journals/aim/19762/ by a University of California San Diego User on 06/03/2017
731
charged from the study when it was apparent that Shigella would not be isolated from an admission stool or rectal swab sample. Eight patients completed one study day; 23, two study days; and three completed 3 days. No patient was given therapy in addition to the study drug. All patients were afebrile at the time of discharge, and of the 23 patients monitored on study day 2, only three (one in each of the study groups) had more than two watery stools on that day.
Adverse Reactions Three patients each in the single-dose and two-dose treatment groups and one patient in the 10-dose group developed signs or symptoms that were possibly related to drug therapy. Two patients had rash and one patient each had tinnitus, swelling of lips, vertigo, generalized arthralgia, and arthritis of the fourth andfifthmetatarsal joints.
Discussion In this study, a single dose of 1 gram of ciprofloxacin, or two doses of 1 gram of ciprofloxacin given on successive days was as effective as conventional 5-day therapy in achieving resolution of clinical symptoms and bacteriologic cure in patients infected with species of Shigella other than S. dysenteriae type 1. Although a placebo-treated control group was not included in this study, previous antimicrobial trials in shigellosis have consistently shown that effective antimicrobial therapy, when compared with placebo or ineffective therapy, significantly shortens the duration and severity of dysentery in Shigella-infected patients (4, 5, 19, 27-29). Thus, the absence of differences among the three treatment groups in this study is attributable to their equivalent efficacy, rather than to spontaneous resolution of symptoms. Short-course ciprofloxacin therapy was less effective than 5-day therapy in the treatment of S. dysenteriae type 1-infected patients. When compared with historical controls who received placebo or ineffective therapy, however, short-course therapy did have an effect on the course of disease. In a previous study of shigellosis using similar entry and outcome criteria, the rate of clinical failure in patients receiving ineffective therapy (ampicillin for treatment of ampicillin-resistant strains of Shigella) was 82.4% (14 of 17 patients) (19). That rate compares with 6 (24%) of 25 S. dysenteriae type 1-infected patients in this study who received short-course therapy (P < 0.001). Surprisingly, in this study, shortcourse therapy was more effective then 5-day therapy in eradicating stool excretion of Shigella and in shortening the duration that mucus was present in stools. The reasons for thisfindingare uncertain, but it may be that the initially higher stool and serum concentrations of the treatment drug provided by the short-course therapy are beneficial. The higher rate of clinical failure among S. dysenteriae type 1-infected patients treated with short-course therapy, when compared with patients infected with 732
other species of Shigella, may be attributable to the at least fourfold lesser activity of ciprofloxacin against S. dysenteriae type 1 when compared with other species of Shigella. This may have allowed for continued replication of Shigella in colonic epithelial cells, where the concentration of ciprofloxacin is presumed to be less than it is in bowel lumen (19), even when Shigella organisms could no longer be recovered from the stool. A previous study by Gottuzo and colleagues (26) found that single-dose therapy with another newer quinolone, norfloxacin, was equivalent to 5 days of therapy with trimethoprim-sulfamethoxazole for the treatment of shigellosis. No patient in that study was infected with S. dysenteriae type 1, however, and most patients had mild disease (a mean of eight stools during the 5-day study period). All participants were treated as outpatients. The results of this study are a reminder that a single disease, such as shigellosis, is heterogeneous, and that new therapies need to be evaluated in diverse groups of patients before definitive recommendations for their use can be made. Virtually all Shigella strains remain susceptible to the newer quinolones, even in areas such as South Asia and Central Africa (6, 19, 20), where resistance to ampicillin, trimethoprim-sulfamethoxazole, and nalidixic acid is common. Thus, a newer quinolone can be given to treat shigellosis with the confidence that it will be effective against the infecting strain. Single-dose therapy and the absence of resistance give the newer quinolones clear practical advantages over treatment with older agents, especially in developing countries where follow-up of patients is often difficult. If patients with shigellosis are treated with an antimicrobial agent to which a substantial proportion of Shigella strains are resistant, the need to re-evaluate patients after 1 or 2 days of therapy can be a considerable burden for already inadequate health services. Compliance with a single-dose regimen is assured if the drug is provided at the health center where the patient is first seen. For travelers to developing countries who are at high risk for contracting Shigella, the convenience of single-dose therapy has obvious advantages. Given these advantages, should single-dose quinolone therapy become the treatment of choice for shigellosis? In developing countries, single-dose quinolone therapy has a number of disadvantages that make it less than an ideal choice. First, the cost of a single 800-mg dose of norfloxacin or 1 gram of ciprofloxacin is $5 to $10 in the United States. Although this is a relatively modest cost for therapy in a developed country, if these drugs were to be priced similarly in developing countries they would be prohibitively expensive for most persons. Second, the newer quinolones are not approved for use in children because of concerns about arthropathy, which has been observed when these drugs were given to juvenile animals (30). In developing countries the need for antimicrobial therapy is greatest among children because the mortality and morbidity from shigellosis is highest in this age group. Nalidixic acid, an older quinolone that is approved for use in children over 3 months of age (31), causes in animal models toxicity similar to that seen with the newer quinolones. Although rarely used in any age group in developed coun-
1 November 1992 • Annals of Internal Medicine • Volume 117 • Number 9
Downloaded From: http://annals.org/pdfaccess.ashx?url=/data/journals/aim/19762/ by a University of California San Diego User on 06/03/2017
tries, nalidixic acid is now being routinely used in children and adults in developing countries for the treatment of shigellosis, and arthropathy has not been found to be a major problem (27). The arthropathy related to the use of newer quinolones appears to be dose-related (32), and patients receiving single-dose therapy would appear to be at minimal risk for developing this complication of therapy. Despite this, there is a reluctance to use newer quinolones in children in developing countries until they have been approved for use in children in developed countries. Because treatment failures occurred with short-course therapy in patients infected with S. dysenteriae type 1, in countries where this species is endemic or epidemic single- or two-dose therapy would not eliminate the need for follow-up evaluation. Facilities for the bacteriologic diagnosis of enteric infections are not routinely available in most of the developing countries where S. dysenteriae type 1 infection is present. Thus patients with suspected shigellosis must often be treated without knowledge of the infecting species of Shigella. It is not possible to determine on clinical grounds which patients with dysentery are infected with S. dysenteriae type 1 (33) and thus who would require follow-up evaluation if they received single- or 2-dose ciprofloxacin therapy. The development of resistance is an additional concern should the newer quinolone be introduced into routine use. Shigella spp. have invariably developed resistance to antimicrobial agents used to treat them. Should the use of newer quinolones in the treatment of shigellosis be restricted to treating infections caused by strains that are resistant to older agents, thus possibly delaying the emergence of resistance to the quinolones? Five-day therapy with amdinocillin pivoxil is currently the treatment of choice in Bangladesh for infections with Shigella strains that are resistant to ampicillin, trimethoprim-sulfamethoxazole, and nalidixic acid (34). Currently, almost all isolates of Shigella in Bangladesh remain susceptible to amdinocillin (35), which is available in a suspension form and which is safe for use in children (amdinocillin, however, is not widely distributed in many countries). Although single-dose quinolone therapy is more convenient than 20-dose amdinocillin therapy (the efficacy of shorter courses of amdinocillin has not been evaluated), from a public health perspective it might be preferable to limit the use of the newer quinolones as long as most Shigella isolates remain susceptible to amdinocillin. Decisions on using short courses of ciprofloxacin or other new quinolone agents for the treatment of shigellosis will differ depending on the patient population and the site where treatment is provided. Given the rapid emergence of multiply resistant strains of Shigella, not only in Asia and Africa but also in Europe and North and South America, it is likely that the newer quinolone agents will be increasingly used to treat Shigella infections. Single-dose therapy with 1 gram of ciprofloxacin is an effective option for treating non-5, dysenteriae type 1 infections. Results from this study suggest that courses of therapy intermediate between 1 and 10 doses would be effective in treating S. dysenteriae type 1 infections, and such courses should be evaluated in larger clinical trials.
Acknowledgments: The authors thank Mr. Humayun Kabir for assistance with data entry, Mr. Md. Rafique Islam Sarker for assistance with laboratory investigations, and the staff of the Diarrhoea Treatment Centre of the International Centre for Diarrhoeal Disease Research, Bangladesh, for patient care. Grant Support: By a grant from Bayer, Leverkusen, Germany, and by the International Centre for Diarrhoeal Disease Research, Bangladesh (ICDDR,B). The ICDDR,B is funded by countries and agencies that share its concern for the health problems of developing countries. Current donors include the aid agencies of the governments of Australia, Bangladesh, Belgium, Canada, Denmark, France, Japan, The Netherlands, Norway, Sweden, Switzerland, the United Kingdom, and the United States; the United Nations Development Program and UNICEF; and the Ford and Sasakawa Foundations. Dr. Bennish is supported by a grant from the Applied Diarrheal Disease Research Project of the United States Agency for International Development. Requests for Reprints: Michael Bennish, MD, New England Medical Center, 750 Washington Street, Box 041, Boston, MA 02111. Current Author Addresses: Dr. Bennish, New England Medical Center, 750 Washington Street, Box 041, Boston, MA 02111. Drs. Salam, W. Khan, and A. Khan: International Centre for Diarrhoeal Disease Research, Bangladesh, GPO Box 128, Dhaka 1000, Bangladesh.
References 1. Bennish ML, Wojtyniak BJ. Mortality due to shigellosis: community and hospital data. Rev Infect Dis. 1991;13(Suppl 4):S245-51. 2. Bennish ML, Harris JR, Wojtyniak BJ, Struelens M. Death in shigellosis: incidence and risk factors in hospitalized patients. J Infect Dis. 1990;161:500-6. 3. Committee on Issues and Priorities for New Vaccine Development. The burden of disease resulting from diarrhea. In: Katz SL; ed. New Vaccine Development: Establishing Priorities. Volume 2. Diseases of Importance in Developing Countries. Washington, DC: National Academy Press; 1986:165. 4. Salam MA, Bennish ML. Antimicrobial therapy for shigellosis. Rev Infect Dis. 1991;13 (Suppl 4):S332-41. 5. Haltalin KC, Nelson JD, Ring R 3d, Sladoje M, Hinton LV. Doubleblind treatment study of shigellosis comparing ampicillin, sulfadiazine, and placebo. J Pediatr. 1967;70:970-81. 6. Bennish ML, Salam MA, Hossain MA, Myaux J, Haque B, Chakraborty J, et al. Antimicrobial resistance among Shigella isolates in Bangladesh, 1983-1990: increasing frequency of strains multiply resistant to ampicillin, trimethoprim-sulfamethoxazole and nalidixic acid. Clin Infec Dis. 1992;14:1055-60. 7. Hyams KC, Bourgeois AL, Merrell BR, Rozmajzl P, Escamilla J, Thornton SA, et al. Diarrheal disease during Operation Desert Shield. N Engl J Med. 1991;325:1423-8. 8. Munshi MH, Sack DA, Haider K, Ahmed ZU, Rahaman MM, Morshed MG. Plasmid-mediated resistance to nalidixic acid in Shigella dysenteriae type 1. Lancet, 1987;2:419-21. 9. Frost JA, Willshaw GA, Barclay EA, Rowe B, Lemmens P, Vandepitte J. Plasmid characterization of drug-resistant Shigella dysenteriae 1 from an epidemic in Central Africa. J of Hyg (Lond). 1985; 94:163-72. 10. Nationwide dissemination of multiply resistant Shigella sonnei following a common source outbreak. MMWR. 1987;36:633-4. 11. Shigella dysenteriae type 1—Guatemala, 1991. MMWR. 1991;40: 421,427-8. 12. Tauxe RV, Puhr ND, Wells JG, Hargrett-Bean N, Blake PA. Antimicrobial resistance of Shigella isolates in the USA: the importance of international travelers. J Infect Dis. 1990;162:1107-11. 13. The rational use of drugs in the management of acute diarrhoea in children. Geneva: World Health Organization; 1990. 14. Lolekha S, Vibulbandhitkit S, Poonyarit P. Response to antimicrobial therapy for shigellosis in Thailand. Rev Infect Dis. 1991; 13(Suppl 4):S342-6. 15. Ronsmans C, Bennish ML, Chakraborty J, Fauveau V. Current practices for treatment of dysentery in rural Bangladesh. Rev Infect Dis. 1991;13(Suppl 4):S351-6. 16. Hossain MM, Glass RI, Khan MR. Antibiotic use in a rural community in Bangladesh. Int J Epidemiol. 1982;11:402-5. 17. Sarder AM, Chen LC. Distribution and characteristics of non-government health practitioners in a rural area of Bangladesh. Soc Sci Med [A]. 1981;15:543-50. 18. Grant JP. The state of the world's children, 1991. New York: Oxford University Press; 1991:112-3. 19. Bennish ML, Salam MA, Haider R, Barza M. Therapy for shigellosis. II. Randomized, double-blind comparison of ciprofloxacin and ampicillin. J Infect Dis. 1990;162:711-6. 20. Ries AA, Wells J, Olivola D, Ntakibirora M, Nyandwi S, Ntibakivayo M, et al. Shigella dysenteriae type 1 infections in Burundi: the end
1 November 1992 • Annals of Internal Medicine • Volume 117 • Number 9
Downloaded From: http://annals.org/pdfaccess.ashx?url=/data/journals/aim/19762/ by a University of California San Diego User on 06/03/2017
733
21.
22.
23.
24.
25. 26.
27.
of the line for antimicrobials. [Abstract 526]. In: Program and abstracts of the 31st Interscience Conference on Antimicrobial Agents and Chemotherapy. Chicago, Illinois. American Society for Microbiology, 1991. Ling J, Kam KM, Lam AW, French GL. Susceptibilities of Hong Kong isolates of multiply resistant Shigella spp. to 25 antimicrobial agents, including ampicillin plus sulbactam and new 4-quinolones. Antimicrob Agents Chemother. 1988;32:20-23. Rogerie F, Ott D, Vandepitte J, Verbist L, Lemmens P, Habiyaremye I. Comparison of norfloxacin and nalidixic acid for treatment of dysentery caused by Shigella dysenteriae type 1 in adults. Antimicrob Agents Chemother. 1986;29:883-6. De Mol P, Mets T, Lagasse R, Vandepitte J, Mutwewingabo A, Butzler JP. Treatment of bacillary dysentery: a comparison between enoxacin and nalidixic acid. J Antimicrob Chemother. 1987;19: 695-8. Davis RL, Koup JR, Williams-Warren J, Weber A, Smith AL. Pharmacokinetics of three oral formulations of ciprofloxacin. Antimicrob Agents Chemother. 1985;28:74-7. Hofken G, Lode H, Prinzing C, Borner K, Koeppe P. Pharmacokinetics of ciprofloxacin after oral and parenteral administration. Antimicrob Agents Chemother. 1985;27:375-9. Gotuzzo E, Oberhelman RA, Maguina C, Berry SJ, Yi A, Guzman M, et al. Comparison of single-dose treatment with norfloxacin and standard 5-day treatment with trimethoprim-sulfamethoxazole for acute shigellosis in adults. Antimicrob Agents Chemother. 1989;33: 1101-4. Salam MA, Bennish ML. Therapy for shigellosis. 1. Randomized
28.
29.
30.
31.
32.
33. 34.
35.
double-blind trial of nalidixic acid in childhood shigellosis. J Pediatr. 1988;113:901-7. Tong MJ, Martin DG, Cunningham JJ, Gunning JJ. Clinical and bacteriological evaluation of antibiotic treatment of shigellosis. JAMA. 1970;214:1841-4. Haltalin KC, Kusmiesz HT, Hinton LV, Nelson JD. Treatment of acute diarrhea in outpatients. Double-blind study comparing ampicillin and placebo. Am J Dis Child. 1972;124;554-61. Ingham B, Brentnall DW, Dale EA, McFadzean JA. Arthropathy induced by antibacterial fused Af-alkyl-4-pyridone-3-carboxylic acids. Toxicol Lett. 1977;1:21-6. Fontaine O. Antibiotics in the management of shigellosis in children: what role for the quinolones? Rev Infect Dis. 1989;ll(Suppl 5): S1145-50. Alfaham M, Holt ME, Goodchild MC. Arthropathy in a child with cystic fibrosis taking ciprofloxacin. Br Med J (Clin Res Ed). 1987; 295:699. Ronsmans C, Bennish ML, Wierzba T. Diagnosis and management of dysentery by community health workers. Lancet. 1988;2:552-5. Kabir I, Rahaman MM, Ahmed SM, Akhter SQ, Butler T. Comparative efficacies of pivmecillinam and ampicillin in acute shigellosis. Antimicrob Agents Chemother. 1984;25:643-5. Mitra AK, Kabir I, Hossain MA. Pivmecillinam-resistant Shigella dysenteriae type 1 infection in Bangladesh [Letter]. Lancet. 1990; 1:1461-2.
The real purpose of the scientific method is to make sure nature hasn't misled you into thinking you know something you don't actually know. Robert M. Pirsig Zen and the Art of Motorcycle Maintenance Submitted by: Mark E. Rupp, MD University of Nebraska Medical Center Omaha, NE 68198
Submissions from readers are welcomed. If the quotation is published, the sender's name will be acknowledged. Please include a complete citation, as done for any reference.—The Editors
734
1 November 1992 • Annals of Internal Medicine • Volume 117 • Number 9
Downloaded From: http://annals.org/pdfaccess.ashx?url=/data/journals/aim/19762/ by a University of California San Diego User on 06/03/2017
• Objective: To determine whether a single dose, or 2 doses, of ciprofloxacin are as effective as 5-day, 10dose therapy for the treatment of shigellosis in adult men who are moderately to severely ill. • Design: Randomized, double-blind clinical trial. • Setting: A diarrhea treatment center in the capital city of a developing country, Bangladesh. • Patients: A total of 128 adult men with dysentery of less than 96 hours duration. All had Shigella organisms isolated from a culture of stool. • Interventions: Patients were randomly assigned to receive either a single 1-gram dose of ciprofloxacin at admission to the study (single-dose group; n = 40), a 1 -gram dose of ciprofloxacin at admission and 24 hours later (2-dose group; n - 43), or 500 mg of ciprofloxacin every 12 hours for 5 days (10 dose group; n = 35). All patients were hospitalized for 6 days. • Measurements: Stools were collected individually; their character and consistency were recorded and cultured daily. A physical examination and recording of symptoms were done daily, and the temperature was measured every 4 hours. Therapy was considered to have failed in patients who did not have improvement in the signs and symptoms of dysentery after 72 hours of therapy or in patients who on study day 5 had more than nine stools, or more than two watery stools, or were febrile. • Results: There were no treatment failures in the 78 patients infected with species of Shigella other than Shigella dysenteriae type 1. Among the 40 patients infected with S. dysenteriae type 1, treatment failed in 4 of the 10 patients who received single-dose therapy, 2 of the 15 patients who received 2-dose therapy, and none of the 15 patients who received 10-dose therapy ( P = 0.017, single-dose therapy group compared with 10-dose group; P = 0.15 for the single-dose group compared with the 2-dose group; P> 0.2 for the 2-dose group compared with the 10-dose group). • Conclusions: A single 1-gram dose of ciprofloxacin is effective therapy for patients infected with species of Shigella other than S. dysenteriae type 1. Single-dose therapy is inferior to 10-dose therapy for treating patients infected with S. dysenteriae type 1.
Annals of Internal Medicine. 1992;117:727-734. From the International Centre for Diarrhoeal Disease Research, Bangladesh, Dhaka, Bangladesh; and New England Medical Center, Boston, Massachusetts. For current author addresses, see end of text.
^Shigellosis remains a major cause of diarrhea-related morbidity and mortality in developing countries (1-3). Antimicrobial therapy is indicated for all persons infected with Shigella spp. who have symptoms of dysentery (4), and effective treatment of such patients reduces resultant morbidity and mortality (4, 5). A number of factors limit the effective therapy of shigellosis. First, in many countries, Shigella strains are frequently resistant to ampicillin, trimethoprim-sulfamethoxazole, and nalidixic acid (6-12), the drugs of choice for treating this infection (4, 13). Resistance is most common among strains of Shigella dysenteriae type 1, which is endemic or epidemic in South and Southeast Asia, Central Africa, and Central America (1, 2, 6, 8, 9, 11, 14). Second, in developing countries, where severe shigellosis is most common, drugs are most often dispensed by persons with minimal—or no—formal medical training (15-17). A full course of therapy is often not prescribed for infections; most commonly, medical practitioners dispense, or prescribe, only a few doses of medication (15, 16). Third, the cost of therapy, especially if newer agents are required, can be prohibitively expensive for persons in developing countries, where the per capita income is often less than $1 per day (18). A short course of ciprofloxacin therapy, if effective, would obviate a number of these problems. First, most multiresistant strains of Shigella remain susceptible to the newer quinolones, including ciprofloxacin (19-21). Second, if a single dose—or two doses—of the drug were effective in treating shigellosis, the failure of most dispensers to provide a 5-day course of therapy would be less of an obstacle to providing effective therapy. Third, patient compliance is likely to be greater than with a conventional 10-dose regimen. Finally, the cost of treatment could be reduced. A short course of ciprofloxacin therapy might be effective in treating shigellosis for several reasons. The newer quinolones, including ciprofloxacin, are known to
SI Unit Conversion Factor Normal Range Blood Leukocyte Count Traditional: 3200 - 9800 mm - 3 mm ~3 x 0.001 SI: 3.2 - 9.8 x 109/L = 109/L Drug Brand Name Generic Name Cipro ciprofloxacin
© 1992 American College of Physicians
Downloaded From: http://annals.org/pdfaccess.ashx?url=/data/journals/aim/19762/ by a University of California San Diego User on 06/03/2017
727
be effective in treating shigellosis when given for 3 to 5 days (14, 19, 22, 23). In relation to the concentration of the drug required to inhibit growth of Shigellae, ciprofloxacin attains high concentrations in serum (mean peak concentrations more than 600 times the minimum inhibitory concentration [MIC] in following a 500-mg dose) and in stool (concentrations uniformly more than 1000 times the MIC) (19). The serum half-life of ciprofloxacin is relatively long, averaging 3 to 5 hours (24, 25). Finally, a single dose of 800 mg of norfloxacin has been found to be effective in treating adults with diarrhea or dysentery of mild severity caused by infection with Shigella flexneri or Shigella sonnei (26). In this study we compared the effectiveness of a single dose of 1 gram of ciprofloxacin, two doses of 1 gram of ciprofloxacin given on successive days, and 500 mg of ciprofloxacin given twice daily for 5 days, on the resolution of symptoms in adult men moderately to severely ill with shigellosis due predominantly to infections with either 5. dysenteriae type 1 or S. flexneri. Methods Patient Population and Selection of Study Patients This study was conducted at the Diarrhoea Treatment Centre of the International Centre for Diarrhoeal Disease Research, Bangladesh, from August 1989 to December, 1990. Patients eligible for study had to fulfill the following criteria:
Table 1. Characteristics of tbte Three 1rreatmentt Groups at Admission* Characteristic Patients, n Age,yt Duration of illness, h Stool frequency before therapy on day of admission^ Vomited before admission Weight, Ag§ Temperature > 37.8° C, admission day, n (%) Mild or moderate dehydration, n(%) Abdominal tenderness, n (%) Shigella sp. isolated, n (%) S. flexneri S. dystenteriae type 1 Other species Other enteric pathogens, n (%) Salmonella sp. Plesiomonas shigelloides or Aeromonas sp. Campylobacter jejuni C. jejuni and Plesiomonas shigelloides Stool microscopic values, n(%) > 50 leukocytes per high power field > 50 erythrocytes per high power field
Treiatment Gi^HE 1-Dose 2-Dose 10-Dose 40 43 35 31 ± 10 26 ± 7 32 ± 10 50 ± 2 0 46 ± 2 1 52 ± 2 0 17 ± 14 21 ± 18 22 ± 16 12 (30) 12 (28) 11 (31) 47 ± 5 50 ± 7 50 ± 9 16 (40)
18 (42)
7(20)
13 (32) 35 (87)
11 (26) 34 (79)
13 (37) 32 (91)
22 (55) 19 (44) 10 (25) 15 (35) 8(20) 9(21)
18 (51) 15 (43) 2(6)
1(3)
1(2)
1(3)
2(3) 0
6(14) 2(5)
2(6) 0
0
0
1(3)
37 (93)
41 (95)
34 (97)
36 (90)
38 (88)
34 (97)
* Values are mean (± SD) unless noted. t P ~ 0.008. One-dose group comf>ared with 2-•dose group, P = 0.02; 2-dose group compared with 10-dose\ group, P =• 0.004. % Determined by history. § P = 0.06. One-dose group compared with 2-dose group., P = 0.04; 1-dose group compared with 10-dosek. group, P »= 0.04. 728
signs and symptoms of dysentery, as manifested by the presence of gross blood and mucus on visual inspection of a stool sample and a history of tenesmus and abdominal pain; male gender; age, 18 to 60 years; illness duration less than 96 hours; no acute illness other than dysentery; and absence of trophozoites of Entamoeba histolytica on microscopic examination of stool. Women were excluded from the study because of concerns about the ability to detect those who were early in a pregnancy and because of social constraints limiting a woman's ability to remain hospitalized for the full 6-day study period. Study Protocol At admission, a complete history and physical examination were done and the following laboratory tests obtained: culture of stool and rectal swab samples for Shigella spp., Salmonella spp., Vibrio spp., Plesiomonas shigelloides, Aeromonas sp., and Campylobacter jejuni; microscopic examination of stool for enumeration of leukocytes and erythrocytes and identification of cysts and ova of parasites; and a complete blood count. All patients were enrolled in the morning and received their first dose of medication at 1400 hours. A study day was considered to begin at 1400 hours and to continue until 1400 hours the next day. Patients were randomly assigned to one of three treatment groups: ciprofloxacin (Cipro; Bayer, Leverkusen, Germany), 500 mg every 12 hours for 5 days (10-dose group); 1 gram of ciprofloxacin given at admission (single-dose group); and 1 gram of ciprofloxacin given at admission and a repeat dose of 1 gram given 24 hours later (two-dose group). Neither the patients nor the staff were aware of which therapy was being given. Medication was provided in blister packages that contained either active drug or placebo, which were identical in appearance and similar in taste. Medication was assigned to a study number using a random number table and a block randomization method with a block size of nine. Patients were hospitalized for 6 full days after the first dose of medication. Symptoms and physical examination findings were recorded daily. Vital signs, including oral temperature, were measured every 4 hours. Fever was defined as an oral temperature greater than 37.8 °C. Stools were collected separately on plastic-lined pads and examined for consistency and the presence of blood or mucus. Stool consistency was graded as watery, soft, or formed according to previously defined criteria (19). Individual stools of varied consistency were categorized according to their predominant characteristic. For each study day, the consistency and character of stools was categorized according to the worst finding in any individual stool sample for that day. A rectal swab or stool sample for culture was obtained daily. Microscopic examination of stool was repeated on study days 3 and 5. Patients were requested to return for evaluation 7 days after discharge. At the return visit a history and physical examination were done, and a stool or rectal swab sample was obtained for culture. Culture and microscopic examination of stool, susceptibility testing, and blood counts were done according to previously described methods (19, 27). Evaluation of Outcome Afebrile patients were considered to have resolution of illness if, on the last day of therapy (study day 5), they had 3 or fewer stools, none watery; to have marked improvement if they had 6 or fewer stools, 1 or fewer watery; to have slight improvement if they had 9 or fewer stools, 2 or fewer watery. Treatment was considered to have failed if, on study day 5, patients were febrile, or had 10 or more stools, or had 3 or more watery stools. Therapy was also considered to have failed in patients who at the end of study day 3 continued to have only frankly bloody-mucoid stools, persistent fever, or severe abdominal pain. All patients in whom therapy failed were treated with ciprofloxacin, 500 mg every 12 hours, for 5 days. Treatment was provided to these patients without knowing which treatment they had previously received. Patients were considered bacteriologically cured if the infecting strain
1 November 1992 • Annals of Internal Medicine • Volume 117 • Number 9
Downloaded From: http://annals.org/pdfaccess.ashx?url=/data/journals/aim/19762/ by a University of California San Diego User on 06/03/2017
of Shigella could not be cultured from a stool or rectal swab sample on study day 3 or thereafter.
Table 2. Clinical Ou tcome by Treatmc;nt Group and Species of Shigella*
Statistical Methods
Shigella Species
Data entry and significance testing were done using a database and statistical program for personal computers (Stat-Pac Gold, version 3.2, Walonick Associates, Minneapolis, Minnesota). Analysis of variance was used to evaluate the significance of differences between the three study groups for continuous variables that were normally distributed; if the difference was significant at P < 0.05, then individual groups were compared using the Students f-test. Differences in continuous variables that were not randomly distributed, such as duration of symptoms, were tested for significance using the Kruskal-Wallis test, and individual groups then were compared using the Mann-Whitney U test if P < 0.05 on the KruskalWallis test. All tests were two-tailed. The significance of differences in proportions was tested using a 2 x 3 chi-square test. If P < 0.05 on the 2 x 3 chi-square test, then individual groups were compared using a 2 x 2 chi-square test, or Fisher exact test if the predicted size of any cell was 5 or less. The initial analysis of the data was blinded; the treatment groups were labeled as 1, 2, and 3 without knowledge of which treatment each group had received. Written informed consent was obtained from all patients. The study protocol was approved by the Ethical and Research Review Committees of the International Centre for Diarrhoeal Disease Research, Bangladesh. Results Patient Enrollment and Admission Characteristics One hundred sixty-two patients were enrolled in the study, 54 in each of the three treatment groups. Thirtyfour (21.0%) patients (10 in the 1-dose group, 8 in the 2-dose group, and 16 in the 10-dose group; P = 0.14) did not have Shigella organisms isolated from an admission rectal swab or stool sample. Ten (7.8%) of the 128 culture-positive patients were excluded from analysis, nine because they withdrew from the study before completion of therapy and one patient because he developed symptomatic rotavirus infection on study day 5. Patients who withdrew from the study did so because they decided not to remain in the hospital for the full study period. Of the nine patients who withdrew from the study, seven did so on the first study day. Four of the 10 excluded patients were in the single-dose group, three in the 2-dose group, and 3 in the 10-dose group (P = 0.2). Thus, of the 154 patients initially entered into the study, 118 patients were eligible for primary analysis comparing the efficacy of the three treatment regimens in patients with culture-confirmed shigellosis. Forty (34%) of these 118 patients were in the singledose treatment group; 43 (36%) in the 2-dose group; and 35 (30%) in the 10-dose group. The admission clinical characteristics of these 118 patients are shown in Table 1. Shigella flexneri (50% of patients) and S. dysenteriae type 1 (34% of patients) were the most common Shigella species isolated from patients. The mean age of patients in the 2-dose treatment group was 26 years, significantly (P < 0.05) less than the mean age of patients in the single-dose and 10-dose groups (mean ages 31 and 32 years, respectively). The three treatment groups did not differ significantly in any other admission characteristic. No patient had clinical signs of malnutrition. Six patients (three in the single-dose group, one in the
Outcoime Resolution Marked Slight Failure Improve- Improvement ment
5(50) 6(40) 5(33)
1(10) 4(27) 8(53)
0 3 (20) 2 (13)
4 (40) 2 (13) 0
24 (80) 25 (89) 18 (90)
6(20) 3(H) 2(10)
0 0 0
0 0 0
29 (72) 31 (72) 23 (66)
7(18) 7(16) 10 (29)
0 3 (7) 2(6)
4 (10) 2 (5) 0
* Clinical outcome was determined oil study day 5. t For S. dysenteriae type$ 1-infected \jatients, 3 )< 4 chi-square test, P = 0.05. Proportion of \>atients in vvhom treatinent failed, 1-dose compared with 10-dose groilp, P = 0.01 7, Fisher ex:act test. Proportion of patients in whom treatnlent failed, 1 -dose and \2-dose groups compared with 10-dose group, iP = 0.05, Fi sher exact itest. % For patients infected with species of Shigella o ther than S. dysenteriae type 1, 3 x 4 chi-sqiiiare test, P > 0.2. § For all patients, 3 x 4 chi-square t 0.:
2-dose group, and two in the 10-dose group), infected with a nalidixic-acid-susceptible strain of Shigella, had received nalidixic acid before study entry; none had received more than four capsules. One patient in the 2-dose group was infected with an ampicillin-susceptible strain of Shigella and had taken four capsules of ampicillin before admission. Six of the seven patients who had previous therapy were infected with S. flexneri, compared with 53 (48%) of the 111 patients who had not received therapy (P = 0.06), and patients who had previous therapy had higher blood leukocyte counts at admission than did patients who had not received therapy (15.2 x 109/L compared with 11.2 x 10 9 /L, P = 0.002). The two groups had no other significant differences in pre-admission characteristics. Excluding the seven patients who had received therapy before entry into the study did not affect the differences among the three treatment groups in any of the major outcome variables.
Clinical Outcome Treatment failed in six patients, four who had received single-dose therapy and 2 who had received 2-dose therapy (Table 2) (P = 0.12, single-dose compared with 10-dose group; P > 0.2, 2-dose compared with 10-dose group). Four of these six patients received open therapy before the end of study treatment because of continued dysenteric symptoms. The remaining two patients in whom treatment failed each had more than 10 stools on the last day of therapy and received open therapy beginning on study day 6. All six patients in whom therapy failed were infected with S. dysenteriae tvne 1. comnared with 34 (30.4%^ of the 112 Datients in
1 November 1992 • Annals of Internal Medicine • Volume 117 • Number 9
Downloaded From: http://annals.org/pdfaccess.ashx?url=/data/journals/aim/19762/ by a University of California San Diego User on 06/03/2017
729
ure 3). Bacteriologic failure occurred in a single patient, who was in the 2-dose therapy group. Shigella dysenteriae type 1 was isolated from a stool sample from this patient on the first three study days, and he was started on open ciprofloxacin therapy on study day 4 because of lack of clinical response. Outcome in Shigella dysenteriae Type 1-Infected Patients
Figure 1. Mean stool frequency by treatment group and study day. The error bars indicate the upper 95% CI of the mean. No statistically significant differences between study groups in the mean number of stools were found for any of the study days. The mean number of stools before admission and on follow-up were determined by history. The pre-admission period is the period from the midnight before admission until admission to the study. For follow-up, the time period was the 24 hours before the visit to the treatment center. For study days 1 to 6, n = 40 in the single-dose group; 43, in the 2-dose group; and 35, in the 10-dose group. At follow-up, n = 34 in the singledose group; 41, in the 2-dose group; and 34, in the 10-dose group. Four patients in the 1-dose group and two patients in each of the 2- and 10-dose groups received open-label, 5-day therapy with ciprofloxacin starting on study day 4 or later. Adm = admission to study; F-up = follow-up.
whom therapy was successful (P = 0.001). Patients in whom therapy failed also had a history of statistically significantly more stools on the day of, and on the day before, admission to the study and had a higher systolic blood pressure when admitted. During the 6 study days, patients in the single-dose group had a mean of 34 stools: those in the 2-dose group, 37 stools; and those in the 10-dose group, 36 stools (P = 0.09) (Figure 1). Only one patient, who was in the single-dose group, was still febrile after 48 hours of therapy (Figure 2). Differences among the study groups in the proportion of patients with abdominal tenderness or tenesmus were not significant for any of the study days. Mucus was present in the stool of patients in the 10-dose group for a longer period (median 6 days) than in patients in the single-dose treatment group (median, four days, P = 0.014) (Table 3). The median duration of other stool characteristics was similar for the three treatment groups. The number of leukocytes and erythrocytes in the stool of the patients in the three treatment groups did not differ significantly on the 3 days (admission, day 3, and day 5) that the test was done.
Bacteriologic Response Shigella organisms were more rapidly eradicated from stool cultures in patients receiving short-course therapy than in those receiving conventional 5-day therapy (Fig730
The 40 patients with S. dysenteriae type 1 disease had more severe illness than did the 78 study patients infected with other species of Shigella. Shigella dysenteriae type 1 infected-patients had significantly (P < 0.05) more stools on each of the 3 days before initiation of the study, and on each of the 6 study days. These patients had a mean of 52 stools during the study, compared with 27 in patients infected with other Shigella species (P < 0.001). Bacterial resolution was also slower in S. dysenteriae type 1 infected-patients; Shigella was isolated in 12 (30%) of 40 5. dysenteriae type 1-infected patients after 1 day of therapy compared with 9 (12%) of 78 patients infected with other Shigella spp. (P = 0.03); after study day 2, the respective figures were seven (18%) compared with two (3%, P = 0.007). All six patients in whom treatment failed and the five patients who had only slight improvement after completion of therapy were among the 40 study patients infected with S. dysenteriae type 1 (see Table 2). Treatment failed in a larger proportion of patients infected with S. dysenteriae type 1 who received single-dose therapy (4 of 10) than in patients who received 10-dose therapy (0 of 15 patients, P = 0.017). The rate of treatment failure in the 2-dose group, 2 of 15 patients, did
Figure 2. Number of patients with fever by treatment group and study day. Fever was defined as an oral temperature greater than 37.8 °C. Differences between the groups were not significant for any of the study days. For study day 1 to 6, n = 4 0 in the single-dose group; 43, in the 2-dose group; and 35, in the 10-dose group. At follow-up, n = 34 in the single-dose group; 41, in the 2-dose group; and 34, in the 10-dose group. Four patients in the 1-dose group and two patients in each of the 2and 10-dose groups received open-label, 5-day therapy with ciprofloxacin starting on study day 4 or later. Adm = admission to study; F-up = follow-up.
1 November 1992 • Annals of Internal Medicine • Volume 117 • Number 9
Downloaded From: http://annals.org/pdfaccess.ashx?url=/data/journals/aim/19762/ by a University of California San Diego User on 06/03/2017
Table 3. Duration c>f Findings by Treatment Group Finding
Treatment Group 1 Dose 2 Dose 10 Dose (n = 40) (n = 43) (n = 35) \jt
0.2). The combined failure rate in the two short-course treatment groups (6 of 25 patients) was significantly (P = 0.05) higher than for patients treated with 10 doses. Shigella dysenteriae type 1-infected patients who failed therapy or who had only slight improvement in symptoms did not differ from other S. dysenteriae type 1-infected patients in duration of illness or stool frequency before therapy. Early in the course of illness, S. dysenteriae type 1-infected patients treated with a single dose of ciprofloxacin had fewer symptoms than did patients treated with conventional 10-dose therapy. After the first day of therapy, patients in the single-dose group less commonly had abdominal pain (6 of 10 patients compared with 15 of 15 patients, P = 0.016), joint pain (2 of 10 patients compared with 10 of 15 patients, P = 0.03), and headache (1 of 10 patients compared with 10 of 15 patients, P = 0.007) than did patients in the 10-dose group. Patients who received 2-dose therapy less frequently had headache or joint pain on the first study day than did the conventional therapy group. On study day 2, anorexia (4 of 10 patients compared with 13 of 15 patients, P = 0.02) and abdominal pain (5 of 10 patients compared with 15 of 15 patients, P = 0.005) were less common in the single-dose group than in the 10-dose group.
of the 78 non-5, dysenteriae type 1 isolates were resistant to ampicillin, 22 (28%) were resistant to trimethoprim-sulfamethoxazole, and only 1 (1%) isolate was resistant to nalidixic acid (P < 0.001 for all three comparisons to S. dysenteriae type 1 isolates). All isolates were susceptible to amdinocillin pivoxil (mecillinam).
Follow-up One-hundred nine of the 118 patients returned for a follow-up evaluation a mean of seven days after discharge from the treatment center. One patient in the single-dose treatment group, who had S. flexneri isolated at admission to the study, but whose stool culture was negative on discharge, had S. flexneri again isolated at follow-up; one other patient in the single-dose group, who had been infected with 5. flexneri at admission to the study, had S. sonnei isolated on his return visit. An additional patient in the single-dose group had recurrent symptoms of dysentery at follow-up; C. jejuni, which was not isolated from an admission or day 5 stool culture, was isolated from a stool culture at follow-up.
Outcome in Patients Not Infected with Shigella Thirty-four patients initially enrolled in the study did not have Shigella isolated from a stool or rectal swab culture. Seven of these patients were infected V. cholera (one with the 0 1 serogroup and six with other serogroups); five with Aeromonas sp., and two each with C. jejuni and P. shigelloides. Patients were dis-
Susceptibility of the Shigella Isolates All Shigella isolates were susceptible to ciprofloxacin when tested by both the disc-diffusion and broth dilution methods. The median MIC of ciprofloxacin for the 35 S. dysenteriae type 1 isolates tested was 0.060 /ig/ mL; for the 59 isolates of other Shigella species tested, it was < 0.004 /ig/mL (P < 0.001). All 40 isolates of S. dysenteriae type 1 were resistant to both ampicillin and trimethoprim-sulfamethoxazole when tested by the disc-diffusion method; 31 (78%) of the 40 isolates were also resistant to nalidixic acid. In comparison, 31 (40%)
Figure 3. Isolation rates of Shigella from stool or rectal swab samples by treatment group and study day. The asterisk indicates that for study day 2, P = 0.04, 2 x 3 chi-square test. P = 0.04, 1-dose compared with the 10-dose group. P = 0.076, 2-dose compared with 10-dose group. For study days 1 to 6, n = 40 in the single-dose group; 43, in the 2-dose group; and 35, in the 10-dose group. At follow-up, n = 34 in the single-dose group; 41, in the 2-dose group; and 34, in the 10-dose group. Four patients in the 1-dose group and two patients in each of the 2- and 10-dose groups received openlabel, 5-day therapy with ciprofloxacin starting on study day 4 or later. Adm = admission to study; F-up = follow up.
1 November 1992 • Annals of Internal Medicine • Volume 117 • Number 9
Downloaded From: http://annals.org/pdfaccess.ashx?url=/data/journals/aim/19762/ by a University of California San Diego User on 06/03/2017
731
charged from the study when it was apparent that Shigella would not be isolated from an admission stool or rectal swab sample. Eight patients completed one study day; 23, two study days; and three completed 3 days. No patient was given therapy in addition to the study drug. All patients were afebrile at the time of discharge, and of the 23 patients monitored on study day 2, only three (one in each of the study groups) had more than two watery stools on that day.
Adverse Reactions Three patients each in the single-dose and two-dose treatment groups and one patient in the 10-dose group developed signs or symptoms that were possibly related to drug therapy. Two patients had rash and one patient each had tinnitus, swelling of lips, vertigo, generalized arthralgia, and arthritis of the fourth andfifthmetatarsal joints.
Discussion In this study, a single dose of 1 gram of ciprofloxacin, or two doses of 1 gram of ciprofloxacin given on successive days was as effective as conventional 5-day therapy in achieving resolution of clinical symptoms and bacteriologic cure in patients infected with species of Shigella other than S. dysenteriae type 1. Although a placebo-treated control group was not included in this study, previous antimicrobial trials in shigellosis have consistently shown that effective antimicrobial therapy, when compared with placebo or ineffective therapy, significantly shortens the duration and severity of dysentery in Shigella-infected patients (4, 5, 19, 27-29). Thus, the absence of differences among the three treatment groups in this study is attributable to their equivalent efficacy, rather than to spontaneous resolution of symptoms. Short-course ciprofloxacin therapy was less effective than 5-day therapy in the treatment of S. dysenteriae type 1-infected patients. When compared with historical controls who received placebo or ineffective therapy, however, short-course therapy did have an effect on the course of disease. In a previous study of shigellosis using similar entry and outcome criteria, the rate of clinical failure in patients receiving ineffective therapy (ampicillin for treatment of ampicillin-resistant strains of Shigella) was 82.4% (14 of 17 patients) (19). That rate compares with 6 (24%) of 25 S. dysenteriae type 1-infected patients in this study who received short-course therapy (P < 0.001). Surprisingly, in this study, shortcourse therapy was more effective then 5-day therapy in eradicating stool excretion of Shigella and in shortening the duration that mucus was present in stools. The reasons for thisfindingare uncertain, but it may be that the initially higher stool and serum concentrations of the treatment drug provided by the short-course therapy are beneficial. The higher rate of clinical failure among S. dysenteriae type 1-infected patients treated with short-course therapy, when compared with patients infected with 732
other species of Shigella, may be attributable to the at least fourfold lesser activity of ciprofloxacin against S. dysenteriae type 1 when compared with other species of Shigella. This may have allowed for continued replication of Shigella in colonic epithelial cells, where the concentration of ciprofloxacin is presumed to be less than it is in bowel lumen (19), even when Shigella organisms could no longer be recovered from the stool. A previous study by Gottuzo and colleagues (26) found that single-dose therapy with another newer quinolone, norfloxacin, was equivalent to 5 days of therapy with trimethoprim-sulfamethoxazole for the treatment of shigellosis. No patient in that study was infected with S. dysenteriae type 1, however, and most patients had mild disease (a mean of eight stools during the 5-day study period). All participants were treated as outpatients. The results of this study are a reminder that a single disease, such as shigellosis, is heterogeneous, and that new therapies need to be evaluated in diverse groups of patients before definitive recommendations for their use can be made. Virtually all Shigella strains remain susceptible to the newer quinolones, even in areas such as South Asia and Central Africa (6, 19, 20), where resistance to ampicillin, trimethoprim-sulfamethoxazole, and nalidixic acid is common. Thus, a newer quinolone can be given to treat shigellosis with the confidence that it will be effective against the infecting strain. Single-dose therapy and the absence of resistance give the newer quinolones clear practical advantages over treatment with older agents, especially in developing countries where follow-up of patients is often difficult. If patients with shigellosis are treated with an antimicrobial agent to which a substantial proportion of Shigella strains are resistant, the need to re-evaluate patients after 1 or 2 days of therapy can be a considerable burden for already inadequate health services. Compliance with a single-dose regimen is assured if the drug is provided at the health center where the patient is first seen. For travelers to developing countries who are at high risk for contracting Shigella, the convenience of single-dose therapy has obvious advantages. Given these advantages, should single-dose quinolone therapy become the treatment of choice for shigellosis? In developing countries, single-dose quinolone therapy has a number of disadvantages that make it less than an ideal choice. First, the cost of a single 800-mg dose of norfloxacin or 1 gram of ciprofloxacin is $5 to $10 in the United States. Although this is a relatively modest cost for therapy in a developed country, if these drugs were to be priced similarly in developing countries they would be prohibitively expensive for most persons. Second, the newer quinolones are not approved for use in children because of concerns about arthropathy, which has been observed when these drugs were given to juvenile animals (30). In developing countries the need for antimicrobial therapy is greatest among children because the mortality and morbidity from shigellosis is highest in this age group. Nalidixic acid, an older quinolone that is approved for use in children over 3 months of age (31), causes in animal models toxicity similar to that seen with the newer quinolones. Although rarely used in any age group in developed coun-
1 November 1992 • Annals of Internal Medicine • Volume 117 • Number 9
Downloaded From: http://annals.org/pdfaccess.ashx?url=/data/journals/aim/19762/ by a University of California San Diego User on 06/03/2017
tries, nalidixic acid is now being routinely used in children and adults in developing countries for the treatment of shigellosis, and arthropathy has not been found to be a major problem (27). The arthropathy related to the use of newer quinolones appears to be dose-related (32), and patients receiving single-dose therapy would appear to be at minimal risk for developing this complication of therapy. Despite this, there is a reluctance to use newer quinolones in children in developing countries until they have been approved for use in children in developed countries. Because treatment failures occurred with short-course therapy in patients infected with S. dysenteriae type 1, in countries where this species is endemic or epidemic single- or two-dose therapy would not eliminate the need for follow-up evaluation. Facilities for the bacteriologic diagnosis of enteric infections are not routinely available in most of the developing countries where S. dysenteriae type 1 infection is present. Thus patients with suspected shigellosis must often be treated without knowledge of the infecting species of Shigella. It is not possible to determine on clinical grounds which patients with dysentery are infected with S. dysenteriae type 1 (33) and thus who would require follow-up evaluation if they received single- or 2-dose ciprofloxacin therapy. The development of resistance is an additional concern should the newer quinolone be introduced into routine use. Shigella spp. have invariably developed resistance to antimicrobial agents used to treat them. Should the use of newer quinolones in the treatment of shigellosis be restricted to treating infections caused by strains that are resistant to older agents, thus possibly delaying the emergence of resistance to the quinolones? Five-day therapy with amdinocillin pivoxil is currently the treatment of choice in Bangladesh for infections with Shigella strains that are resistant to ampicillin, trimethoprim-sulfamethoxazole, and nalidixic acid (34). Currently, almost all isolates of Shigella in Bangladesh remain susceptible to amdinocillin (35), which is available in a suspension form and which is safe for use in children (amdinocillin, however, is not widely distributed in many countries). Although single-dose quinolone therapy is more convenient than 20-dose amdinocillin therapy (the efficacy of shorter courses of amdinocillin has not been evaluated), from a public health perspective it might be preferable to limit the use of the newer quinolones as long as most Shigella isolates remain susceptible to amdinocillin. Decisions on using short courses of ciprofloxacin or other new quinolone agents for the treatment of shigellosis will differ depending on the patient population and the site where treatment is provided. Given the rapid emergence of multiply resistant strains of Shigella, not only in Asia and Africa but also in Europe and North and South America, it is likely that the newer quinolone agents will be increasingly used to treat Shigella infections. Single-dose therapy with 1 gram of ciprofloxacin is an effective option for treating non-5, dysenteriae type 1 infections. Results from this study suggest that courses of therapy intermediate between 1 and 10 doses would be effective in treating S. dysenteriae type 1 infections, and such courses should be evaluated in larger clinical trials.
Acknowledgments: The authors thank Mr. Humayun Kabir for assistance with data entry, Mr. Md. Rafique Islam Sarker for assistance with laboratory investigations, and the staff of the Diarrhoea Treatment Centre of the International Centre for Diarrhoeal Disease Research, Bangladesh, for patient care. Grant Support: By a grant from Bayer, Leverkusen, Germany, and by the International Centre for Diarrhoeal Disease Research, Bangladesh (ICDDR,B). The ICDDR,B is funded by countries and agencies that share its concern for the health problems of developing countries. Current donors include the aid agencies of the governments of Australia, Bangladesh, Belgium, Canada, Denmark, France, Japan, The Netherlands, Norway, Sweden, Switzerland, the United Kingdom, and the United States; the United Nations Development Program and UNICEF; and the Ford and Sasakawa Foundations. Dr. Bennish is supported by a grant from the Applied Diarrheal Disease Research Project of the United States Agency for International Development. Requests for Reprints: Michael Bennish, MD, New England Medical Center, 750 Washington Street, Box 041, Boston, MA 02111. Current Author Addresses: Dr. Bennish, New England Medical Center, 750 Washington Street, Box 041, Boston, MA 02111. Drs. Salam, W. Khan, and A. Khan: International Centre for Diarrhoeal Disease Research, Bangladesh, GPO Box 128, Dhaka 1000, Bangladesh.
References 1. Bennish ML, Wojtyniak BJ. Mortality due to shigellosis: community and hospital data. Rev Infect Dis. 1991;13(Suppl 4):S245-51. 2. Bennish ML, Harris JR, Wojtyniak BJ, Struelens M. Death in shigellosis: incidence and risk factors in hospitalized patients. J Infect Dis. 1990;161:500-6. 3. Committee on Issues and Priorities for New Vaccine Development. The burden of disease resulting from diarrhea. In: Katz SL; ed. New Vaccine Development: Establishing Priorities. Volume 2. Diseases of Importance in Developing Countries. Washington, DC: National Academy Press; 1986:165. 4. Salam MA, Bennish ML. Antimicrobial therapy for shigellosis. Rev Infect Dis. 1991;13 (Suppl 4):S332-41. 5. Haltalin KC, Nelson JD, Ring R 3d, Sladoje M, Hinton LV. Doubleblind treatment study of shigellosis comparing ampicillin, sulfadiazine, and placebo. J Pediatr. 1967;70:970-81. 6. Bennish ML, Salam MA, Hossain MA, Myaux J, Haque B, Chakraborty J, et al. Antimicrobial resistance among Shigella isolates in Bangladesh, 1983-1990: increasing frequency of strains multiply resistant to ampicillin, trimethoprim-sulfamethoxazole and nalidixic acid. Clin Infec Dis. 1992;14:1055-60. 7. Hyams KC, Bourgeois AL, Merrell BR, Rozmajzl P, Escamilla J, Thornton SA, et al. Diarrheal disease during Operation Desert Shield. N Engl J Med. 1991;325:1423-8. 8. Munshi MH, Sack DA, Haider K, Ahmed ZU, Rahaman MM, Morshed MG. Plasmid-mediated resistance to nalidixic acid in Shigella dysenteriae type 1. Lancet, 1987;2:419-21. 9. Frost JA, Willshaw GA, Barclay EA, Rowe B, Lemmens P, Vandepitte J. Plasmid characterization of drug-resistant Shigella dysenteriae 1 from an epidemic in Central Africa. J of Hyg (Lond). 1985; 94:163-72. 10. Nationwide dissemination of multiply resistant Shigella sonnei following a common source outbreak. MMWR. 1987;36:633-4. 11. Shigella dysenteriae type 1—Guatemala, 1991. MMWR. 1991;40: 421,427-8. 12. Tauxe RV, Puhr ND, Wells JG, Hargrett-Bean N, Blake PA. Antimicrobial resistance of Shigella isolates in the USA: the importance of international travelers. J Infect Dis. 1990;162:1107-11. 13. The rational use of drugs in the management of acute diarrhoea in children. Geneva: World Health Organization; 1990. 14. Lolekha S, Vibulbandhitkit S, Poonyarit P. Response to antimicrobial therapy for shigellosis in Thailand. Rev Infect Dis. 1991; 13(Suppl 4):S342-6. 15. Ronsmans C, Bennish ML, Chakraborty J, Fauveau V. Current practices for treatment of dysentery in rural Bangladesh. Rev Infect Dis. 1991;13(Suppl 4):S351-6. 16. Hossain MM, Glass RI, Khan MR. Antibiotic use in a rural community in Bangladesh. Int J Epidemiol. 1982;11:402-5. 17. Sarder AM, Chen LC. Distribution and characteristics of non-government health practitioners in a rural area of Bangladesh. Soc Sci Med [A]. 1981;15:543-50. 18. Grant JP. The state of the world's children, 1991. New York: Oxford University Press; 1991:112-3. 19. Bennish ML, Salam MA, Haider R, Barza M. Therapy for shigellosis. II. Randomized, double-blind comparison of ciprofloxacin and ampicillin. J Infect Dis. 1990;162:711-6. 20. Ries AA, Wells J, Olivola D, Ntakibirora M, Nyandwi S, Ntibakivayo M, et al. Shigella dysenteriae type 1 infections in Burundi: the end
1 November 1992 • Annals of Internal Medicine • Volume 117 • Number 9
Downloaded From: http://annals.org/pdfaccess.ashx?url=/data/journals/aim/19762/ by a University of California San Diego User on 06/03/2017
733
21.
22.
23.
24.
25. 26.
27.
of the line for antimicrobials. [Abstract 526]. In: Program and abstracts of the 31st Interscience Conference on Antimicrobial Agents and Chemotherapy. Chicago, Illinois. American Society for Microbiology, 1991. Ling J, Kam KM, Lam AW, French GL. Susceptibilities of Hong Kong isolates of multiply resistant Shigella spp. to 25 antimicrobial agents, including ampicillin plus sulbactam and new 4-quinolones. Antimicrob Agents Chemother. 1988;32:20-23. Rogerie F, Ott D, Vandepitte J, Verbist L, Lemmens P, Habiyaremye I. Comparison of norfloxacin and nalidixic acid for treatment of dysentery caused by Shigella dysenteriae type 1 in adults. Antimicrob Agents Chemother. 1986;29:883-6. De Mol P, Mets T, Lagasse R, Vandepitte J, Mutwewingabo A, Butzler JP. Treatment of bacillary dysentery: a comparison between enoxacin and nalidixic acid. J Antimicrob Chemother. 1987;19: 695-8. Davis RL, Koup JR, Williams-Warren J, Weber A, Smith AL. Pharmacokinetics of three oral formulations of ciprofloxacin. Antimicrob Agents Chemother. 1985;28:74-7. Hofken G, Lode H, Prinzing C, Borner K, Koeppe P. Pharmacokinetics of ciprofloxacin after oral and parenteral administration. Antimicrob Agents Chemother. 1985;27:375-9. Gotuzzo E, Oberhelman RA, Maguina C, Berry SJ, Yi A, Guzman M, et al. Comparison of single-dose treatment with norfloxacin and standard 5-day treatment with trimethoprim-sulfamethoxazole for acute shigellosis in adults. Antimicrob Agents Chemother. 1989;33: 1101-4. Salam MA, Bennish ML. Therapy for shigellosis. 1. Randomized
28.
29.
30.
31.
32.
33. 34.
35.
double-blind trial of nalidixic acid in childhood shigellosis. J Pediatr. 1988;113:901-7. Tong MJ, Martin DG, Cunningham JJ, Gunning JJ. Clinical and bacteriological evaluation of antibiotic treatment of shigellosis. JAMA. 1970;214:1841-4. Haltalin KC, Kusmiesz HT, Hinton LV, Nelson JD. Treatment of acute diarrhea in outpatients. Double-blind study comparing ampicillin and placebo. Am J Dis Child. 1972;124;554-61. Ingham B, Brentnall DW, Dale EA, McFadzean JA. Arthropathy induced by antibacterial fused Af-alkyl-4-pyridone-3-carboxylic acids. Toxicol Lett. 1977;1:21-6. Fontaine O. Antibiotics in the management of shigellosis in children: what role for the quinolones? Rev Infect Dis. 1989;ll(Suppl 5): S1145-50. Alfaham M, Holt ME, Goodchild MC. Arthropathy in a child with cystic fibrosis taking ciprofloxacin. Br Med J (Clin Res Ed). 1987; 295:699. Ronsmans C, Bennish ML, Wierzba T. Diagnosis and management of dysentery by community health workers. Lancet. 1988;2:552-5. Kabir I, Rahaman MM, Ahmed SM, Akhter SQ, Butler T. Comparative efficacies of pivmecillinam and ampicillin in acute shigellosis. Antimicrob Agents Chemother. 1984;25:643-5. Mitra AK, Kabir I, Hossain MA. Pivmecillinam-resistant Shigella dysenteriae type 1 infection in Bangladesh [Letter]. Lancet. 1990; 1:1461-2.
The real purpose of the scientific method is to make sure nature hasn't misled you into thinking you know something you don't actually know. Robert M. Pirsig Zen and the Art of Motorcycle Maintenance Submitted by: Mark E. Rupp, MD University of Nebraska Medical Center Omaha, NE 68198
Submissions from readers are welcomed. If the quotation is published, the sender's name will be acknowledged. Please include a complete citation, as done for any reference.—The Editors
734
1 November 1992 • Annals of Internal Medicine • Volume 117 • Number 9
Downloaded From: http://annals.org/pdfaccess.ashx?url=/data/journals/aim/19762/ by a University of California San Diego User on 06/03/2017
