Aliment. Pharmacol. Ther. (1992) 6, 727-737.
Maintenance therapy in duodenal and gastric ulcer disease : survey of practice amongst British gas troenterologists
E.J. S. BOYD, J. G. P E N S T O N & K. G. WORMSLEY Ninewells Hospital and Medical School, Dundee, UK Accepted for publication 6 July 1992
SUMMARY
We have used a postal questionnaire to obtain data on the practice of maintenance therapy for peptic ulcer disease by members of the British Society of Gastroenterology. Completed questionnaires were returned by 434 members. Ninety-six per cent used maintenance therapy for patients with duodenal ulcer and 81% for gastric ulcer. Maintenance therapy was considered to be safe (duodenal ulcer 91%; gastric ulcer 78%), acceptable to patients (duodenal ulcer gastric ulcer 89%; gastric ulcer 80%)and to reduce the incidence of ulcer complications (duodenal ulcer 81 %; gastric ulcer 68 YO).There was consensus that increasjng age of patient, current use of non-steroidal anti-inflammatory drugs, previous ulcer complications, and ulcer relapse after surgery were relatively strong indications for maintenance therapy. However, the proportion of patients who received maintenance therapy varied widely amongst respondents (from < 10% to > 50%). There was no agreement on the optimal duration of therapy, nor on management of patients who relapsed during maintenance therapy. It appears that the criteria for use of maintenance therapy need to be better defined, and that established knowledge about the practice of maintenance therapy should be better disseminated and acted upon. Correspondence to : Dr K. G. Wormsley, Ninewells Hospital and Medical School, Dundee DDI 9SY, UK. 72 7
728
E. J. S. BOYD et al.
INTRODUCTION The chronic relapsing course and propensity to develop complications of peptic ulcer disease pose considerable medical and socio-economic problems. One of the therapeutic strategies which attempts to solve some of these problems is maintenance therapy, in which anti-ulcer medication is administered continuously to prevent ulcer recurrence and complications. Many studies of different aspects of maintenance therapy have been undertaken, and many contradictory recommendations have been made concerning the selection of patients for, and clinical practice of, maintenance therapy. Because no information is available about how the published studies and recommendations have influenced gastroenterological practice, we have undertaken a survey to determine current attitudes to and use of maintenance therapy by members of the British Society of Gastroenterology. The objectives were to define general points of agreement and disagreement in the practice of maintenance therapy, in an attempt to provide therapeutic guidelines based on consensus and to identify questions which require further evaluation. METHODS
A questionnaire was sent to all ordinary members of the British Society of Gastroenterology whose entries in the Members’ Handbook indicated that they might be involved in the therapeutic management of patients with peptic ulcer disease. Members whose entries indicated that they were pathologists, radiologists, paediatricians, non-clinical scientists or members of the pharmaceutical industry were not included in the survey. The questionnaire ensured the anonymity of the respondents but obtained information about their speciality (medicine or surgery), and whether they held University or National Health Service posts. The questionnaire has assessed opinions about five principal topics relating to maintenance therapy. (I) Impression about overall safety, acceptability to patients, and effectiveness in reducing the complications of peptic ulcer disease. (2) Whether respondents use maintenance therapy and, if so, the proportion of ulcer patients under the care of the respondent who receive maintenance therapy. ( 3 ) Patient characteristics which influence the decision to use maintenance therapy. (4) Ulcer characteristics which influence the decision to use maintenance therapy. (5) Clinical practice of maintenance therapy, including choice of drug, followup of patients and management of relapse. With the exception of the choice of drug, questions required either a yes/no response, or were in the form of a stem question followed by a range of answer options of which only one was to be chosen. When a respondent failed to indicate
SURVEY OF M A I N T E N A N C E THERAPY
729
Table I. Reasons for failure to complete returned questionnaires Non-clinical speciality/do not encounter ulcers in practice Refer to colleagues for medical treatment Retired from medical practice Moved or on sabbatical No reason given
21
I1 10 6 4
a choice, or indicated more than one option, the reply was included in analysis as a ’don’t know’ response. Replies to questions concerning the safety and acceptability to patients of maintenance therapy have been expressed as percentages of all respondents (that is, whether or not they used maintenance therapy). Replies to subsequent questions relating to the selection of patients and the practice of maintenance therapy have been expressed as percentages of respondents who use maintenance therapy for that particular indication (for example, duodenal ulcer or gastric ulcer). In reply to the request to list the three drugs most commonly used for maintenance therapy in the order of frequency in which they were prescribed, some respondents only listed one or two drugs. Replies have therefore been expressed as the number of times a given drug was mentioned as a percentage of the total number of drug selections for each of the first, second and third choices. In comparisons between subgroups of respondents (for example, between test was used, combining cells when statistically surgeons and physicians) the appropriate. In all tables percentages have been expressed as the nearest whole number.
xz
RESULTS Questionnaires were sent to 731 members and returned by 486 (66%).Fifty-two members did not complete the questionnaire for a variety of reasons (Table 1).The respondents comprised 239 National Health Service physicians, 106 National Health Service surgeons, 47 University physicians, 2 7 University surgeons, and 15 whose speciality and/or affiliation was not specified.
Safety, acceptability and use of maintenance therapy The majority of respondents considered that maintenance therapy was safe, acceptable to patients and reduced the incidence of complications. The proportions holding these views were higher in the case of duodenal ulcer than in gastric ulcer (Table 2). Maintenance therapy was used by more respondents for duodenal ulcer than for gastric ulcer. The proportions of patients treated with maintenance therapy showed a wide spread amongst respondents (Figure I).
730
E. J. S. B O Y D ef al.
Table 2. Use of maintenance therapy: safety, acceptability to patients, effects on complications amongst respondents
Use maintenance Yes No Don’t know Safe Yes No Don’t know Acceptable to patients Yes No Don’t know Reduces complications Yes No Don’t know
-Duodenal
ulcer-
Duodenal ulcer (n = 434)
Gastric ulcer (n = 434)
417 (96%) 14 (3%) 3 (1Yo)
350 (81yo) 8 1 (19%) 3 (1%)
397 (91Yo) 13 (3%) 24 (6%)
338 (78%) 54 (12YO) 44 (10%)
387 (89%) 23 (5%) 24 (6%)
347 (80%) 49 (11%) 38 (9%)
354 (81Yo) 5 1 (12%) 29 (7%)
297 (68%) 93 (21%) 44 (10%)
-Gastric
ulcer-
21-30%
I
0
I
10
I
I
I
l
l
I
20
30
0
10
20
30
Percentage of users ( n =417)
Percentage of users ( n =350)
Figure 1. Use of maintenance therapy. Respondents were asked whether they used maintenance therapy in duodenal ulcer and in gastric ulcer. If the answer was ‘yes’ respondents were asked to indicate in approximately what proportion of patients (0-10%; 11-20%; 21-30%; 31-50%; or > 50%). (Ordinate: percentage of all patients receiving maintenance therapy; abscissa: percentage of respondents in each category.)
Patient characteristics influencing the decision to use rnainfenance therapy There was consensus that continuous treatment with non-steroidal antiinflammatory drugs (NSAIDs),anticoagulants or glucocorticoids, the presence of serious systemic disease, and increasing age were indications for maintenance therapy in both duodenal and gastric ulcer disease (Table 3). Patient gender and smoking habits were not considered positive indications by the majority of respondents (Table 3 ) .
S U R V E Y OF MAINTENANCE THERAPY
731
Table 3. Patient criteria influencing decision to use maintenance therapy Duodenal ulcer (n = 417)
Gastric ulcer (n = 350)
400 (96%) 16 (4%) 1 (1%)
334 (95 %) 14 (4%) 2 (1%)
312 (75%) 100 (24%) 5 (1%)
261 (75%) 83 (24%) 6 (1%)
325 (78%) 83 (20%) 9 (2 Yo)
275 (79%) 63 (18%) 12 (3%)
319 (76%) 97 (23%) 1 (1%)
267 (76 Yo) 79 (23%) 4 (1%)
293 (70%) 120 (29%) 4 (1%)
248 (71%) 95 (27%) 7 (2%)
50 (12%) 360 (86%) 7 (2%)
30 (9%) 313 (89%) 7 (2%)
198 (47%) 215 (52%) 4 (1%)
147 (42%) 194 (55 %) 9 (3 Yo)
__
NSAIDs Yes No Don’t know Glucocorticoids Yes No Don’t know Anticoagulants Yes No Don’t know Systemic Disease Yes No Don’t know Age Yes No Don’t know Gender Yes No Don’t know Smoking habits Yes No Don’t know
Ulcer characteristics influencing the decision to use maintenance therapy Frequency of symptomatic relapses (although there was no consensus on the minimum frequency of relapse which constituted an indication), a history of previous ulcer haemorrhage or perforation, and the occurrence of relapse after previous attempted curative surgery were considered to be indications for maintenance therapy by the majority of respondents (Table 4). Respondents used maintenance therapy in a higher proportion of patients with a previous history of ulcer haemorrhage, perforation or recurrence after surgery than they did in patients overall (Figure 2 compared with Figure I).Indeed, onethird to one-half of all respondents used maintenance therapy in 75-100% of patients under these circumstances.
732
E. J. S. B O Y D et al.
Table 4. Ulcer characteristics influencing the decision to use maintenance therapy Duodenal ulcer Gastric ulcer ( n = 350) (n = 417) ~-
~~
Duration of history Yes
No Don't know
315 (76%) 96 (23 %) 6 (I%)
241 (69%) 101 (29%) 8 (2%)
31 (10%) 237 (75 %) 33 (10%) 5 (2%) 9 (3 %)
54 (22%) 151 (63%) 24 (10%) 4 (2%) 8 (3%)
390 (94 %) 24 (6%) 3 (1%)
295 (84%) 49 (14%) 6 (2 %)
135 (35%) 116 (30%) 133 (34%) 6 (2%)
126 (43Yo) 95 (32%) 67 (23 %) 7 (2 %)
356 (85 %) 56 (13%) 5 (1%)
297 (85 %) 47 (13%) 6 (2 %)
306 (73%) 103 (25 %) 8 (2%)
245 (70%) 90 (26%) 15 (4%)
386 (93%) 27 (6%) 4 (1Yo)
308 (88%) 33 (9%) 9 (3%)
If yes, minimum duration
< 1 year 2-5 years 6-10 years > 10 years Don't know Frequency of Relapse Yes
No Don't know
If yes, minimum frequency
< I year-' 1-2 years-' 3 years--' Don't know Previous Haemorrhage Yes
>
No Don't know Previous Perforation Yes
No Don't know Relapse after Surgery Yes
No Don't know
Clinical Practice of maintenance therapy The majority of respondents considered it necessary to confirm gastric ulcer healing endoscopically (Table 5). Opinion was evenly divided on whether patients with gastric ulcer disease should be followed-up in an outpatient clinic. Neither endoscopic confirmation of healing nor outpatient follow-up were considered necessary in duodenal ulcer disease by the majority of respondents. In spite of the tendency not to follow up patients in clinics, most respondents considered that symptoms suggestive of ulcer recurrence warranted endoscopic re-examination in both duodenal and gastric ulcer disease. Periodic endoscopic examination to detect asymptomatic recurrences was not undertaken by the majority of respondents.
SURVEY OF M A I N T E N A N C E THERAPY Duodenal ulcer
Figure 2. Use of maintenance therapy in duodenal ulcer or gastric ulcer after previous ulcer complications or relapse after attempted curative surgery. Respondents were asked to indicate whether they used maintenance therapy after the listed ulcer complications or when relapse occurred after attempted curative surgery. If the answer was ’yes’ they were asked in what proportion of patients (025%; 25-50%; 51-75%; 76100%). (Ordinate: percentage of all patients receiving maintenance therapy; abscissa: percentage of respondents in each category.)
-
a,
Gastric ulcer
Hoemorrhoge
I
Don’t known
c
B
& c
Perforation
s
0-25%?
E 2
v)
+
x6
+
26-50%=7 5 l - 7 0 % m
76-100%-
I
Don’t k n o w l
r
Relapse after surgery
Don’t know
1 I
0
I I
I
I
I
I
I
I
10 20 30 40 50 60 Respondents (%)
0
I
I
Confirm healing Yes
No Don’t know Out-patient follow-up Yes
No Don’t know Endoscope for symptoms Yes
No Don’t know Endoscope asymptomatic patients Yes Don’t know
I
I
I
I
10 20 30 40 50 60 Respondents
Table 5. Clinical practice of maintenance therapy
No
733
Duodenal ulcer (n = 417)
(n = 350)
Gastric ulcer
119 (29%) 295 (71%) 3 (1%)
269 (77yo) 76 (22%) 5 (2%)
141 (34%) 264 (63 %) 12 (3%)
160 (46%) 181 (52%) 9 (3%)
313 (75 %) 95 (23%) 9 (2%)
307 (88%) 33 (9%) 10 (3Yo)
25 (6%) 373 (89%) 19 (5%)
79 (23 Yo) 259 (74%) 12 (3 %)
P/o)
734
E. J. S. BOYD ef al.
Table 6. Duration of maintenance therapy
1 year 2-5 years 6-10 years > 10 years
Don’t know
Duodenal ulcer (n = 417)
Gastric ulcer (n = 350)
90 (22%) 116 (28%) 29 (7%) 138 (33%) 44 (11%)
79 (23%) 81 (23%) 25 (7%) 120 (34%) 45 (13%)
Table 7. Usual management of second symptomatic relapse during maintenance therapy Duodenal ulcer (n = 417) Reheal and return to standard maintenance dose Reheal and return to increased maintenance dose Change to alternative medical therapy Refer for surgery Don’t know
Gastric ulcer (n = 350)
62 (15 %)
48 (14%)
110 (26%)
98 (28%)
138 (33 %)
64 (18%)
61 (15%) 46 (11%)
101 (29%) 39 (11%)
There was no consensus of opinion on the duration of maintenance therapy (Table 6). Similarly, opinion varied concerning the management of patients who had suffered a second symptomatic relapse during maintenance therapy (that is, who had failed to respond to standard-dose maintenance therapy) (Table 7). In both duodenal and gastric ulcers, rehealing and increasing the dose of maintenance drug was the usual management of one-quarter of respondents, while a slightly greater proportion indicated either a change of treatment for patients with duodenal ulcer, or surgery for gastric ulcer. The drugs most commonly indicated as being the first and second choice for maintenance therapy were ranitidine and cimetidine respectively. Omeprazole was the drug most commonly stated as being third choice. Mucosal protective agents were seldom used.
Differences in practice of maintenance therapy between physicians and surgeons Fewer surgeons than physicians considered maintenance therapy to be safe, acceptable, and to reduce the incidence of complications in gastric ulcer disease, and fewer used maintenance therapy for this indication (Table 8).However, in both duodenal and gastric ulcer disease, surgeons who did use maintenance therapy used it in a higher proportion of their patients than did physicians (for duodenal ulcer: 40% us. 29%, P < 0.001; for gastric ulcer: 37% us. 28%, P = 0.0028).
SURVEY OF M A I N T E N A N C E T H E R A P Y
735
Table 8. Use of maintenance therapy: safety, acceptability to patients and effects on complications in relation to medical or surgical speciality Gastric ulcer
Duodenal ulcer -~
Use maintenance Safe Acceptable to patients Reduces complications
Surgeons (n = 133)
Physicians (n = 286)
Surgeons ( n = 133)
Physicians ( n = 286)
125 (94%) 109 (82%)' 105 (79%) 97 (73 %)
278 274 258 243
91 (68%)" 75 (56%)* 89 (67%)" 64 (48%)'
247 (86%) 251 (88%) 246 (86%) 221 (77%)
(97%) (96%) (90%) (85%)
' Indicates statistically significant difference between surgeons and physicians. Table 9. Management of second symptomatic relapse during maintenance therapy in relation to medical or surgical specialityt Duodenal uIcer Surgeons (n = 133) Reheal and return to standard maintenance dose Reheal and return to increased maintenance dose Change to alternative medical therapy Refer for surgery
Gastric ulcer Physicians ( n = 286)
Surgeons (n = 133)
Physicians ( n = 286)
14 (11%)
48 (17%)
10 (8%)
38 (13%)
23 (17%)"
85 (30%)
11 (8%)"
85 (30%)
26 (20 %)*
107 (37%)
41 (31%)*
48 (17%)
41 (31%)"
18 (6%)
38 (29%)"
47 (16%)
* Indicates statistically significant difference between surgeons and physicians.
t Non-users and 'don't knows' excluded. In both duodenal and gastric ulcer surgeons were significantly more likely than physicians to recommend that maintenance therapy be given for only one year (for duodenal ulcer: 36% us. 19%, P = 0.0007; for gastric ulcer: 41% us. 21%, P = 0.0016), while physicians were significantly more likely to recommend that maintenance therapy be continued for longer periods. In the management of duodenal or gastric ulcers which relapse symptomatically for a second time during maintenance therapy surgeons were significantly more likely to recommend surgical treatment, whereas physicians recommend either an increase in the dose of maintenance drug or, in the case of duodenal ulcer, a change in therapy (for example, eradication of Helicobacfer pylori) (Table 9). The only significant difference in maintenance practice between respondents in academic posts and those in NHS posts was that the former were significantly more likely to undertake outpatient follow-up of both duodenal and gastric ulcer patients.
736
E. J. S. BOYD ef al.
Duodenal ulcer
1st (n=4161
2nd (n=3591
3rd ( n =I601
Ranitidine Cimetidine Sucralfate Misoprostol Denol Antacids Famotidine Nizati d ine Omeprazole Caved-S
Gastric ulcer
0 10 20 30 4 0 50 60 0 10 20 30 40 50 60 0 10 20 30 40 50 60 Ist ( n=343)
2nd (n=311)
3rd ( n =I351
Ranitidine Cimetidine Sucralfate Misoprostol Denol Antacids Farnotidine Nizat idine Omeprazole Caved-S
0 10 20 30 40 50 60 0 10 20 30 40 50 60 0 10 20 30 40 50 60 Choices (%) Choices (%I Choices (%) Figure 3. Choice of drugs used for maintenance therapy. Respondents were asked to list the three drugs which they most commonly used for maintenance therapy in order of decreasing frequency of use.
DISCUSSION This study has provided extensive new data on how British gastroenterologists use maintenance therapy in duodenal and gastric ulcer disease. The majority of respondents considered maintenance therapy to be safe and acceptable to patients, and also to decrease the incidence of ulcer complications. If the overall objectives of peptic ulcer therapy are to keep patients free from ulcer symptoms and complications without causing unwanted treatment-related effects, then clearly maintenance therapy is perceived to fulfill these objectives. However, the proportion of patients who received maintenance therapy varied widely among British gastroenterologists. Overall, only about one-third of the patients with either duodenal ulcer or gastric ulcer received maintenance therapy. The use of maintenance therapy in only a minority of ulcer patients, despite the perceived effectiveness and safety, may reflect preference for other therapeutic options (for example, eradication therapy for H. pylovi or curative surgery) or a belief that maintenance therapy is either not necessary, or unacceptably expensive, for the
SURVEY OF M A I N T E N A N C E THERAPY
737
majority of patients. The reasons and justification for particular therapeutic choices were not assessed by the questionnaire. One of the striking findings of this survey was the diversity of opinion among respondents about the criteria for selecting patients for maintenance therapy and the conduct of maintenance therapy, so that there was little consensus on which to base therapeutic recommendations. O n the other hand, on some of the topics where there was consensus, the uniform attitude did not appear to be justifiable on the basis of scientific evidence. For example, the consensus view that maintenance therapy is indicated in elderly patients, and in patients who have suffered previously from ulcer complications or whose uIcers have recurred after surgery, is strongly supported by clinical evidence, since these patients are at an increased risk of further complications and have an increased mortality from the ulcer disease. However, although most respondents agreed that continuous treatment with ulcerogenic drugs was an indication for maintenance therapy, the latter appears to be an empirical judgement, since there are no satisfactory controlled clinical studies either for or against such an indication. Conversely, poor agreement between respondents sometimes seemed to reflect discrepancy between the theoretical recommendations for maintenance therapy (based on the results of controlled investigations), and the practical use of maintenance therapy. For example, many studies have shown that stopping maintenance therapy after 1year, or even after 5 years, results in relapse of the ulcer disease-yet nearly half of the respondents used maintenance therapy for less than 5 years. It is thus possible to draw two important conclusions from the present study. It seems that although we have described many of the clinical features which determine the practice of maintenance treatment by British gastroenterologists, it is still not possible to define the causes of the striking variations in practice both between gastroenterologists and amongst their patients. Consequently, it appears that much of the knowledge concerning maintenance therapy which has already been obtained and consolidated by well-designed clinical investigation has either not been adequately disseminated or is not acted upon. The explanation for this therapeutic conundrum is not obvious and represents an excellent topic for clinical audit. ACKNOWLEDGEMENTS
We should like to thank all those members of the British Society of Gastroenterology who completed and returned the questionnaires. We are grateful to Miss F. MacDonald for her assistance in addressing and mailing the questionnaires.
38
BAP 6
Maintenance therapy in duodenal and gastric ulcer disease : survey of practice amongst British gas troenterologists
E.J. S. BOYD, J. G. P E N S T O N & K. G. WORMSLEY Ninewells Hospital and Medical School, Dundee, UK Accepted for publication 6 July 1992
SUMMARY
We have used a postal questionnaire to obtain data on the practice of maintenance therapy for peptic ulcer disease by members of the British Society of Gastroenterology. Completed questionnaires were returned by 434 members. Ninety-six per cent used maintenance therapy for patients with duodenal ulcer and 81% for gastric ulcer. Maintenance therapy was considered to be safe (duodenal ulcer 91%; gastric ulcer 78%), acceptable to patients (duodenal ulcer gastric ulcer 89%; gastric ulcer 80%)and to reduce the incidence of ulcer complications (duodenal ulcer 81 %; gastric ulcer 68 YO).There was consensus that increasjng age of patient, current use of non-steroidal anti-inflammatory drugs, previous ulcer complications, and ulcer relapse after surgery were relatively strong indications for maintenance therapy. However, the proportion of patients who received maintenance therapy varied widely amongst respondents (from < 10% to > 50%). There was no agreement on the optimal duration of therapy, nor on management of patients who relapsed during maintenance therapy. It appears that the criteria for use of maintenance therapy need to be better defined, and that established knowledge about the practice of maintenance therapy should be better disseminated and acted upon. Correspondence to : Dr K. G. Wormsley, Ninewells Hospital and Medical School, Dundee DDI 9SY, UK. 72 7
728
E. J. S. BOYD et al.
INTRODUCTION The chronic relapsing course and propensity to develop complications of peptic ulcer disease pose considerable medical and socio-economic problems. One of the therapeutic strategies which attempts to solve some of these problems is maintenance therapy, in which anti-ulcer medication is administered continuously to prevent ulcer recurrence and complications. Many studies of different aspects of maintenance therapy have been undertaken, and many contradictory recommendations have been made concerning the selection of patients for, and clinical practice of, maintenance therapy. Because no information is available about how the published studies and recommendations have influenced gastroenterological practice, we have undertaken a survey to determine current attitudes to and use of maintenance therapy by members of the British Society of Gastroenterology. The objectives were to define general points of agreement and disagreement in the practice of maintenance therapy, in an attempt to provide therapeutic guidelines based on consensus and to identify questions which require further evaluation. METHODS
A questionnaire was sent to all ordinary members of the British Society of Gastroenterology whose entries in the Members’ Handbook indicated that they might be involved in the therapeutic management of patients with peptic ulcer disease. Members whose entries indicated that they were pathologists, radiologists, paediatricians, non-clinical scientists or members of the pharmaceutical industry were not included in the survey. The questionnaire ensured the anonymity of the respondents but obtained information about their speciality (medicine or surgery), and whether they held University or National Health Service posts. The questionnaire has assessed opinions about five principal topics relating to maintenance therapy. (I) Impression about overall safety, acceptability to patients, and effectiveness in reducing the complications of peptic ulcer disease. (2) Whether respondents use maintenance therapy and, if so, the proportion of ulcer patients under the care of the respondent who receive maintenance therapy. ( 3 ) Patient characteristics which influence the decision to use maintenance therapy. (4) Ulcer characteristics which influence the decision to use maintenance therapy. (5) Clinical practice of maintenance therapy, including choice of drug, followup of patients and management of relapse. With the exception of the choice of drug, questions required either a yes/no response, or were in the form of a stem question followed by a range of answer options of which only one was to be chosen. When a respondent failed to indicate
SURVEY OF M A I N T E N A N C E THERAPY
729
Table I. Reasons for failure to complete returned questionnaires Non-clinical speciality/do not encounter ulcers in practice Refer to colleagues for medical treatment Retired from medical practice Moved or on sabbatical No reason given
21
I1 10 6 4
a choice, or indicated more than one option, the reply was included in analysis as a ’don’t know’ response. Replies to questions concerning the safety and acceptability to patients of maintenance therapy have been expressed as percentages of all respondents (that is, whether or not they used maintenance therapy). Replies to subsequent questions relating to the selection of patients and the practice of maintenance therapy have been expressed as percentages of respondents who use maintenance therapy for that particular indication (for example, duodenal ulcer or gastric ulcer). In reply to the request to list the three drugs most commonly used for maintenance therapy in the order of frequency in which they were prescribed, some respondents only listed one or two drugs. Replies have therefore been expressed as the number of times a given drug was mentioned as a percentage of the total number of drug selections for each of the first, second and third choices. In comparisons between subgroups of respondents (for example, between test was used, combining cells when statistically surgeons and physicians) the appropriate. In all tables percentages have been expressed as the nearest whole number.
xz
RESULTS Questionnaires were sent to 731 members and returned by 486 (66%).Fifty-two members did not complete the questionnaire for a variety of reasons (Table 1).The respondents comprised 239 National Health Service physicians, 106 National Health Service surgeons, 47 University physicians, 2 7 University surgeons, and 15 whose speciality and/or affiliation was not specified.
Safety, acceptability and use of maintenance therapy The majority of respondents considered that maintenance therapy was safe, acceptable to patients and reduced the incidence of complications. The proportions holding these views were higher in the case of duodenal ulcer than in gastric ulcer (Table 2). Maintenance therapy was used by more respondents for duodenal ulcer than for gastric ulcer. The proportions of patients treated with maintenance therapy showed a wide spread amongst respondents (Figure I).
730
E. J. S. B O Y D ef al.
Table 2. Use of maintenance therapy: safety, acceptability to patients, effects on complications amongst respondents
Use maintenance Yes No Don’t know Safe Yes No Don’t know Acceptable to patients Yes No Don’t know Reduces complications Yes No Don’t know
-Duodenal
ulcer-
Duodenal ulcer (n = 434)
Gastric ulcer (n = 434)
417 (96%) 14 (3%) 3 (1Yo)
350 (81yo) 8 1 (19%) 3 (1%)
397 (91Yo) 13 (3%) 24 (6%)
338 (78%) 54 (12YO) 44 (10%)
387 (89%) 23 (5%) 24 (6%)
347 (80%) 49 (11%) 38 (9%)
354 (81Yo) 5 1 (12%) 29 (7%)
297 (68%) 93 (21%) 44 (10%)
-Gastric
ulcer-
21-30%
I
0
I
10
I
I
I
l
l
I
20
30
0
10
20
30
Percentage of users ( n =417)
Percentage of users ( n =350)
Figure 1. Use of maintenance therapy. Respondents were asked whether they used maintenance therapy in duodenal ulcer and in gastric ulcer. If the answer was ‘yes’ respondents were asked to indicate in approximately what proportion of patients (0-10%; 11-20%; 21-30%; 31-50%; or > 50%). (Ordinate: percentage of all patients receiving maintenance therapy; abscissa: percentage of respondents in each category.)
Patient characteristics influencing the decision to use rnainfenance therapy There was consensus that continuous treatment with non-steroidal antiinflammatory drugs (NSAIDs),anticoagulants or glucocorticoids, the presence of serious systemic disease, and increasing age were indications for maintenance therapy in both duodenal and gastric ulcer disease (Table 3). Patient gender and smoking habits were not considered positive indications by the majority of respondents (Table 3 ) .
S U R V E Y OF MAINTENANCE THERAPY
731
Table 3. Patient criteria influencing decision to use maintenance therapy Duodenal ulcer (n = 417)
Gastric ulcer (n = 350)
400 (96%) 16 (4%) 1 (1%)
334 (95 %) 14 (4%) 2 (1%)
312 (75%) 100 (24%) 5 (1%)
261 (75%) 83 (24%) 6 (1%)
325 (78%) 83 (20%) 9 (2 Yo)
275 (79%) 63 (18%) 12 (3%)
319 (76%) 97 (23%) 1 (1%)
267 (76 Yo) 79 (23%) 4 (1%)
293 (70%) 120 (29%) 4 (1%)
248 (71%) 95 (27%) 7 (2%)
50 (12%) 360 (86%) 7 (2%)
30 (9%) 313 (89%) 7 (2%)
198 (47%) 215 (52%) 4 (1%)
147 (42%) 194 (55 %) 9 (3 Yo)
__
NSAIDs Yes No Don’t know Glucocorticoids Yes No Don’t know Anticoagulants Yes No Don’t know Systemic Disease Yes No Don’t know Age Yes No Don’t know Gender Yes No Don’t know Smoking habits Yes No Don’t know
Ulcer characteristics influencing the decision to use maintenance therapy Frequency of symptomatic relapses (although there was no consensus on the minimum frequency of relapse which constituted an indication), a history of previous ulcer haemorrhage or perforation, and the occurrence of relapse after previous attempted curative surgery were considered to be indications for maintenance therapy by the majority of respondents (Table 4). Respondents used maintenance therapy in a higher proportion of patients with a previous history of ulcer haemorrhage, perforation or recurrence after surgery than they did in patients overall (Figure 2 compared with Figure I).Indeed, onethird to one-half of all respondents used maintenance therapy in 75-100% of patients under these circumstances.
732
E. J. S. B O Y D et al.
Table 4. Ulcer characteristics influencing the decision to use maintenance therapy Duodenal ulcer Gastric ulcer ( n = 350) (n = 417) ~-
~~
Duration of history Yes
No Don't know
315 (76%) 96 (23 %) 6 (I%)
241 (69%) 101 (29%) 8 (2%)
31 (10%) 237 (75 %) 33 (10%) 5 (2%) 9 (3 %)
54 (22%) 151 (63%) 24 (10%) 4 (2%) 8 (3%)
390 (94 %) 24 (6%) 3 (1%)
295 (84%) 49 (14%) 6 (2 %)
135 (35%) 116 (30%) 133 (34%) 6 (2%)
126 (43Yo) 95 (32%) 67 (23 %) 7 (2 %)
356 (85 %) 56 (13%) 5 (1%)
297 (85 %) 47 (13%) 6 (2 %)
306 (73%) 103 (25 %) 8 (2%)
245 (70%) 90 (26%) 15 (4%)
386 (93%) 27 (6%) 4 (1Yo)
308 (88%) 33 (9%) 9 (3%)
If yes, minimum duration
< 1 year 2-5 years 6-10 years > 10 years Don't know Frequency of Relapse Yes
No Don't know
If yes, minimum frequency
< I year-' 1-2 years-' 3 years--' Don't know Previous Haemorrhage Yes
>
No Don't know Previous Perforation Yes
No Don't know Relapse after Surgery Yes
No Don't know
Clinical Practice of maintenance therapy The majority of respondents considered it necessary to confirm gastric ulcer healing endoscopically (Table 5). Opinion was evenly divided on whether patients with gastric ulcer disease should be followed-up in an outpatient clinic. Neither endoscopic confirmation of healing nor outpatient follow-up were considered necessary in duodenal ulcer disease by the majority of respondents. In spite of the tendency not to follow up patients in clinics, most respondents considered that symptoms suggestive of ulcer recurrence warranted endoscopic re-examination in both duodenal and gastric ulcer disease. Periodic endoscopic examination to detect asymptomatic recurrences was not undertaken by the majority of respondents.
SURVEY OF M A I N T E N A N C E THERAPY Duodenal ulcer
Figure 2. Use of maintenance therapy in duodenal ulcer or gastric ulcer after previous ulcer complications or relapse after attempted curative surgery. Respondents were asked to indicate whether they used maintenance therapy after the listed ulcer complications or when relapse occurred after attempted curative surgery. If the answer was ’yes’ they were asked in what proportion of patients (025%; 25-50%; 51-75%; 76100%). (Ordinate: percentage of all patients receiving maintenance therapy; abscissa: percentage of respondents in each category.)
-
a,
Gastric ulcer
Hoemorrhoge
I
Don’t known
c
B
& c
Perforation
s
0-25%?
E 2
v)
+
x6
+
26-50%=7 5 l - 7 0 % m
76-100%-
I
Don’t k n o w l
r
Relapse after surgery
Don’t know
1 I
0
I I
I
I
I
I
I
I
10 20 30 40 50 60 Respondents (%)
0
I
I
Confirm healing Yes
No Don’t know Out-patient follow-up Yes
No Don’t know Endoscope for symptoms Yes
No Don’t know Endoscope asymptomatic patients Yes Don’t know
I
I
I
I
10 20 30 40 50 60 Respondents
Table 5. Clinical practice of maintenance therapy
No
733
Duodenal ulcer (n = 417)
(n = 350)
Gastric ulcer
119 (29%) 295 (71%) 3 (1%)
269 (77yo) 76 (22%) 5 (2%)
141 (34%) 264 (63 %) 12 (3%)
160 (46%) 181 (52%) 9 (3%)
313 (75 %) 95 (23%) 9 (2%)
307 (88%) 33 (9%) 10 (3Yo)
25 (6%) 373 (89%) 19 (5%)
79 (23 Yo) 259 (74%) 12 (3 %)
P/o)
734
E. J. S. BOYD ef al.
Table 6. Duration of maintenance therapy
1 year 2-5 years 6-10 years > 10 years
Don’t know
Duodenal ulcer (n = 417)
Gastric ulcer (n = 350)
90 (22%) 116 (28%) 29 (7%) 138 (33%) 44 (11%)
79 (23%) 81 (23%) 25 (7%) 120 (34%) 45 (13%)
Table 7. Usual management of second symptomatic relapse during maintenance therapy Duodenal ulcer (n = 417) Reheal and return to standard maintenance dose Reheal and return to increased maintenance dose Change to alternative medical therapy Refer for surgery Don’t know
Gastric ulcer (n = 350)
62 (15 %)
48 (14%)
110 (26%)
98 (28%)
138 (33 %)
64 (18%)
61 (15%) 46 (11%)
101 (29%) 39 (11%)
There was no consensus of opinion on the duration of maintenance therapy (Table 6). Similarly, opinion varied concerning the management of patients who had suffered a second symptomatic relapse during maintenance therapy (that is, who had failed to respond to standard-dose maintenance therapy) (Table 7). In both duodenal and gastric ulcers, rehealing and increasing the dose of maintenance drug was the usual management of one-quarter of respondents, while a slightly greater proportion indicated either a change of treatment for patients with duodenal ulcer, or surgery for gastric ulcer. The drugs most commonly indicated as being the first and second choice for maintenance therapy were ranitidine and cimetidine respectively. Omeprazole was the drug most commonly stated as being third choice. Mucosal protective agents were seldom used.
Differences in practice of maintenance therapy between physicians and surgeons Fewer surgeons than physicians considered maintenance therapy to be safe, acceptable, and to reduce the incidence of complications in gastric ulcer disease, and fewer used maintenance therapy for this indication (Table 8).However, in both duodenal and gastric ulcer disease, surgeons who did use maintenance therapy used it in a higher proportion of their patients than did physicians (for duodenal ulcer: 40% us. 29%, P < 0.001; for gastric ulcer: 37% us. 28%, P = 0.0028).
SURVEY OF M A I N T E N A N C E T H E R A P Y
735
Table 8. Use of maintenance therapy: safety, acceptability to patients and effects on complications in relation to medical or surgical speciality Gastric ulcer
Duodenal ulcer -~
Use maintenance Safe Acceptable to patients Reduces complications
Surgeons (n = 133)
Physicians (n = 286)
Surgeons ( n = 133)
Physicians ( n = 286)
125 (94%) 109 (82%)' 105 (79%) 97 (73 %)
278 274 258 243
91 (68%)" 75 (56%)* 89 (67%)" 64 (48%)'
247 (86%) 251 (88%) 246 (86%) 221 (77%)
(97%) (96%) (90%) (85%)
' Indicates statistically significant difference between surgeons and physicians. Table 9. Management of second symptomatic relapse during maintenance therapy in relation to medical or surgical specialityt Duodenal uIcer Surgeons (n = 133) Reheal and return to standard maintenance dose Reheal and return to increased maintenance dose Change to alternative medical therapy Refer for surgery
Gastric ulcer Physicians ( n = 286)
Surgeons (n = 133)
Physicians ( n = 286)
14 (11%)
48 (17%)
10 (8%)
38 (13%)
23 (17%)"
85 (30%)
11 (8%)"
85 (30%)
26 (20 %)*
107 (37%)
41 (31%)*
48 (17%)
41 (31%)"
18 (6%)
38 (29%)"
47 (16%)
* Indicates statistically significant difference between surgeons and physicians.
t Non-users and 'don't knows' excluded. In both duodenal and gastric ulcer surgeons were significantly more likely than physicians to recommend that maintenance therapy be given for only one year (for duodenal ulcer: 36% us. 19%, P = 0.0007; for gastric ulcer: 41% us. 21%, P = 0.0016), while physicians were significantly more likely to recommend that maintenance therapy be continued for longer periods. In the management of duodenal or gastric ulcers which relapse symptomatically for a second time during maintenance therapy surgeons were significantly more likely to recommend surgical treatment, whereas physicians recommend either an increase in the dose of maintenance drug or, in the case of duodenal ulcer, a change in therapy (for example, eradication of Helicobacfer pylori) (Table 9). The only significant difference in maintenance practice between respondents in academic posts and those in NHS posts was that the former were significantly more likely to undertake outpatient follow-up of both duodenal and gastric ulcer patients.
736
E. J. S. BOYD ef al.
Duodenal ulcer
1st (n=4161
2nd (n=3591
3rd ( n =I601
Ranitidine Cimetidine Sucralfate Misoprostol Denol Antacids Famotidine Nizati d ine Omeprazole Caved-S
Gastric ulcer
0 10 20 30 4 0 50 60 0 10 20 30 40 50 60 0 10 20 30 40 50 60 Ist ( n=343)
2nd (n=311)
3rd ( n =I351
Ranitidine Cimetidine Sucralfate Misoprostol Denol Antacids Farnotidine Nizat idine Omeprazole Caved-S
0 10 20 30 40 50 60 0 10 20 30 40 50 60 0 10 20 30 40 50 60 Choices (%) Choices (%I Choices (%) Figure 3. Choice of drugs used for maintenance therapy. Respondents were asked to list the three drugs which they most commonly used for maintenance therapy in order of decreasing frequency of use.
DISCUSSION This study has provided extensive new data on how British gastroenterologists use maintenance therapy in duodenal and gastric ulcer disease. The majority of respondents considered maintenance therapy to be safe and acceptable to patients, and also to decrease the incidence of ulcer complications. If the overall objectives of peptic ulcer therapy are to keep patients free from ulcer symptoms and complications without causing unwanted treatment-related effects, then clearly maintenance therapy is perceived to fulfill these objectives. However, the proportion of patients who received maintenance therapy varied widely among British gastroenterologists. Overall, only about one-third of the patients with either duodenal ulcer or gastric ulcer received maintenance therapy. The use of maintenance therapy in only a minority of ulcer patients, despite the perceived effectiveness and safety, may reflect preference for other therapeutic options (for example, eradication therapy for H. pylovi or curative surgery) or a belief that maintenance therapy is either not necessary, or unacceptably expensive, for the
SURVEY OF M A I N T E N A N C E THERAPY
737
majority of patients. The reasons and justification for particular therapeutic choices were not assessed by the questionnaire. One of the striking findings of this survey was the diversity of opinion among respondents about the criteria for selecting patients for maintenance therapy and the conduct of maintenance therapy, so that there was little consensus on which to base therapeutic recommendations. O n the other hand, on some of the topics where there was consensus, the uniform attitude did not appear to be justifiable on the basis of scientific evidence. For example, the consensus view that maintenance therapy is indicated in elderly patients, and in patients who have suffered previously from ulcer complications or whose uIcers have recurred after surgery, is strongly supported by clinical evidence, since these patients are at an increased risk of further complications and have an increased mortality from the ulcer disease. However, although most respondents agreed that continuous treatment with ulcerogenic drugs was an indication for maintenance therapy, the latter appears to be an empirical judgement, since there are no satisfactory controlled clinical studies either for or against such an indication. Conversely, poor agreement between respondents sometimes seemed to reflect discrepancy between the theoretical recommendations for maintenance therapy (based on the results of controlled investigations), and the practical use of maintenance therapy. For example, many studies have shown that stopping maintenance therapy after 1year, or even after 5 years, results in relapse of the ulcer disease-yet nearly half of the respondents used maintenance therapy for less than 5 years. It is thus possible to draw two important conclusions from the present study. It seems that although we have described many of the clinical features which determine the practice of maintenance treatment by British gastroenterologists, it is still not possible to define the causes of the striking variations in practice both between gastroenterologists and amongst their patients. Consequently, it appears that much of the knowledge concerning maintenance therapy which has already been obtained and consolidated by well-designed clinical investigation has either not been adequately disseminated or is not acted upon. The explanation for this therapeutic conundrum is not obvious and represents an excellent topic for clinical audit. ACKNOWLEDGEMENTS
We should like to thank all those members of the British Society of Gastroenterology who completed and returned the questionnaires. We are grateful to Miss F. MacDonald for her assistance in addressing and mailing the questionnaires.
38
BAP 6
