The Laryngoscope C 2015 The American Laryngological, V

Rhinological and Otological Society, Inc.

Diagnostic Accuracy of History and Physical Examination in Bacterial Acute Rhinosinusitis Timo J. Autio, MD; Timo Koskenkorva, MD; Mervi N€ arki€o, MD; Tuomo K. Leino, MD; Petri Koivunen, MD; Olli-Pekka Alho, MD Objectives/Hypothesis: To evaluate the diagnostic accuracy of symptoms, the symptom progression pattern, and clinical signs in identifying bacterial acute rhinosinusitis (ARS). Study Design: We conducted an inception cohort study among 50 military recruits with ARS. Methods: We collected symptoms daily from the onset of symptoms to approximately 10 days. At 9 to 10 days, standardized data on symptoms and physical findings were gathered. A positive culture of maxillary sinus aspirate was considered to be the reference standard for bacterial ARS. Results: At 9 to 10 days, the presence or deterioration after 5 days of any of the symptoms could not be used to diagnose bacterial ARS. Toothache had an adequate positive likelihood ratio (positive likelihood ratio [LR1] 4.4) but was too rare to be used for screening. In contrast, several physical findings at 9 to 10 days were of more diagnostic use and frequent enough for screening. Moderate or profuse (vs. none/minimal) amount of secretion in nasal passage seen in anterior rhinoscopy satisfactorily either ruled in, if present (LR1 3.2), or ruled out, if absent (negative likelihood ratio 0.2), bacterial ARS. If any secretion was seen in the posterior pharynx or middle meatus, the probability of bacterial ARS increased markedly (LR1 5.3 and LR1 11.0, respectively). Conclusion: We found symptoms or their change to be of little use in identifying bacterial ARS. In contrast, we observed several clinical findings after 9 to 10 days of symptoms to predict bacterial ARS quite accurately. Key Words: Acute rhinosinusitis, bacterial, diagnosis, clinical signs, symptoms, sensitivity, specificity, likelihood ratio, physical examination, cohort study. Level of Evidence: 4. Laryngoscope, 125:1541–1546, 2015

INTRODUCTION Antibiotics are often used for acute rhinosinusitis (ARS),1,2 yet most episodes are caused by viral infection.3,4 Overconsumption of antibiotics is related to antibiotic resistance and has an influence on normal bacterial flora.5,6 Even in normal use, diarrhea, abdominal pain, and vomiting may follow. Therefore, avoiding unnecessary antibiotic treatment is crucial. According to European and American guidelines3,4 we should diagnose and treat bacterial ARS, which should be differentiated from viral ARS primarily by symptoms and clinical findings. This is hampered by the fact that the same symptoms and findings are common in both bacterial and viral ARS.3 Unsurprisingly, recent Cochrane reviews

From the Department of Otorhinolaryngology and Medical Research Center Oulu, Oulu University Hospital and University of Oulu (T.J.A., T.K., P.K., O–P.A.), Oulu; Centre for Military Medicine (M.N.), and the Air Force Command Finland (T.K.L.), Finnish Defence Forces, Finland Editor’s Note: This Manuscript was accepted for publication February 10, 2015. The authors received a grant for material and laboratory costs from the Finnish Defense Forces. The authors have no other funding, financial relationships, or conflicts of interest to disclose. Send correspondence to Timo J. Autio, MD, Department of Otorhinolaryngology, Institute of Clinical Medicine, University of Oulu, P.O. Box 5000, FI-90014 University of Oulu, Finland. E-mail: [email protected] DOI: 10.1002/lary.25247

Laryngoscope 125: July 2015

found the effect of antibiotic treatment to be questionable in uncomplicated ARS diagnosed on the basis of clinical criteria.7,8 More information is thus needed on the diagnostic accuracy of different symptoms, the symptom progression pattern, and clinical signs in identifying bacterial ARS. Antimicrobials only have an effect on bacterial ARS, so the reference criterion must be culture-proven bacterial ARS.9 Three meta-analyses9–11 found no diagnostic studies on ARS where such a reference standard was primarily used. Moreover, only four studies12–15 were found that used the appearance of sinus punctate as the reference criterion, and these studies had several methodological defects (e.g., selected patient material, unsatisfactory description of eligibility criteria, patient characteristics and symptoms, lack of blinding). Most importantly, these were cross-sectional studies in which the diagnostic tests and reference outcome were measured at the same visit, causing inevitable recall bias. We hypothesized that the use of a prospective follow-up, with daily collection of symptoms; standardized recording of symptoms and physical findings; and—most importantly—a maxillary sinus puncture and bacterial culture as the reference standard, would provide more accurate information on clinical criteria that are valid for the diagnosis of bacterial ARS. Thus, we conducted an inception cohort study among military recruits during a single episode of ARS and explored the Autio et al.: Diagnosis of Bacterial Acute Rhinosinusitis

1541

diagnostic accuracy of various symptoms and physical findings in bacterial ARS.

MATERIALS AND METHODS Study Population and Protocol We conducted a prospective diagnostic test study (inception cohort) among military recruits carrying out compulsory military service in the Kajaani Garrison, Kainuu Brigade, Northern Finland. The patients were recruited and examined at the garrison health center (primary care). All the patients provided written informed consent, and the study protocol was approved by the Ethics Committee of the Northern Ostrobothnian Hospital District. The study protocol was registered in the ClinicalTrials.gov database (No. NCT01580137). The participants were enrolled from consecutive recruits who sought medical care because of acute respiratory symptoms between February 1 and April 15, 2012. The participants had to have ARS based on the following inclusion criteria: acute onset within the preceding 4 days, presence of nasal symptoms (blockage or discharge),3 and abnormal nasal findings (mucosal edema or secretion). Exclusion criteria were concomitant infection requiring antimicrobial treatment, respiratory infection or antimicrobial treatment within 3 weeks preceding the first visit, nasal allergy or asthma necessitating medication, chronic nasal symptoms or polyps, and prior nasal or sinus surgery. At enrollment, we collected background information and gave the participants symptom diaries. Viral polymerase chain reaction (PCR) samples were taken, as described earlier,16 and a follow-up visit was scheduled on the workday closest to the 10th day of symptoms. The 10th day was chosen according to recent guidelines.3,4 Only oral paracetamol (1000 mg x 1–3 per day) was allowed for symptom relief. No irrigation, mist, or any nasal medication was allowed.

Symptoms and Clinical Examination Recording of Symptoms. The participants recorded the presence of acute symptoms and graded their severity from 0 (none) to 10 (worst possible) in the symptom diary daily for 10 days, as suggested by the European guidelines on rhinosinusitis.3 The following symptoms were documented: nasal symptoms (nasal blockage, clear nasal discharge, purulent nasal discharge, postnasal discharge, reduction of smell), pain symptoms (facial pain/pressure, headache, otalgia, toothache), and other symptoms (sleeping disorder, fatigue, snoring, cough). The patients filled in the symptom diary from memory for the days they had had symptoms prior to enrollment so that the diary started from the onset of symptoms. Definition of Presence of Symptoms and Pattern of Symptom Progression. At the follow-up visit, a score  1 in the symptom diary for that day indicated that the subject had that symptom. We also tested the cutoff points of a score  4 (moderate symptoms present) or  8 (severe symptoms present), but these yielded essentially similar results (data not shown). Any increase in the score after 5 days was considered a worsening symptom (“double-sickening”).4 The increase was recorded for each symptom. We also evaluated the accuracy of worsening of symptoms by using a cutoff value of 3 days, but this also produced essentially similar results (data not shown). Physical Examination. At the follow-up day, one otolaryngologist (T.J.A.) used a headlight, a tongue depressor, and a nasal speculum to grade clinical findings according to the standardized method described earlier.17 The color of the nasal mucosa was classified as normal or erythematous. Nasal

Laryngoscope 125: July 2015

1542

mucosal swelling was graded as mild (inferior turbinate prominent but middle turbinate visible), moderate (middle turbinate obscured but inferior turbinate not touching the septum), or severe (inferior turbinate touching the septum). The color (clear vs. colored—white, yellow, green), consistency (thin vs. thick vs. crusty), and amount (none/minimal vs. moderate vs. profuse) of the nasal secretion was evaluated. Any secretion in the posterior pharynx (yes vs. no), tenderness in the face (yes vs. no), or cervical adenopathy (yes vs. no) was recorded. Finally, we used a 2.7-mm 30-degree rigid nasal endoscope (TrueView 160 mm telescope, Olympus Europa, Hamburg, Germany) with an EndoLED (Olympus, Europa) light source to look for any secretion in the middle meatus (yes vs. no). All examinations were done blinded to the symptom scores and radiological and microbiological findings. Reference Standard. A positive culture of maxillary sinus aspirate with a known respiratory pathogen was considered to be the reference standard for bacterial ARS.4 On the follow-up day (the workday closest to the 10th day of symptoms), the subjects were chosen for a maxillary sinus puncture according to the following criteria. First, we performed a paranasal cone-beam computed tomography (CBCT) for all the patients with an open cone-beam three-dimensional (3D) system (Scanora 3D, Soredex Inc, Tuusula, Finland). We used standard-resolution 8-mA current and small-sized fields of view (height 6.0 cm, diameter 6.0 cm) with a voxel size of 0.20 mm. If we saw an air-fluid level, gas bubbles, or total opacification in the maxillary sinus in the CBCT scan, we punctured that sinus. The puncture was done through the inferior meatus under topical anesthesia (solution of lidocaine 40 mg/ml and adrenalin 0.05 mg/ml and lidocaine spray 10 mg/dos) with a 2-0 G needle. We applied 2 to 3 ml of sterile sodium chloride solution into the sinus and aspirated the contents with a 10-ml syringe, avoiding contamination. We then injected the aspirates into a Portagerm bottle (BıoM erıeux SA, France). The aspirates were analyzed for both aerobic and anaerobic bacterial cultures on sheep blood, chocolated blood, and fastidious anaerobic blood-agar plates in aerobic and anaerobic atmospheres at 35 C for 48 hours to find common respiratory bacterial pathogens causing ARS. All pathogens were identified at the species level. The cultures were confirmed with broad-range bacterial PCR, as described earlier.16

Statistical Analyses For descriptive data, we calculated means with ranges or medians with interquartile (IQR) ranges, depending on the normality of the data. Associations between culture-proven bacterial ARS and various symptoms and clinical findings were studied further by cross-tabulation and calculation of sensitivity, specificity, 95% confidence intervals (CI), and positive and negative likelihood ratios (LR1 and LR–). All the data were analyzed per patient. The worst side was chosen when the nasal clinical findings were evaluated. If the status variable included over two groups, the cutoff value was chosen based on which variable values had the highest specificity to find bacterial ARS.

RESULTS Study Population Of the 66 patients offered participation, 51 consented and were enrolled. One patient with pneumonia was excluded. Seven (14%) patients who did not return the diary were excluded from the analyses of symptoms. The mean age of the participants was 20 years (range Autio et al.: Diagnosis of Bacterial Acute Rhinosinusitis

TABLE I. Diagnostic Accuracy of the Presence of Various Symptoms Recorded at 9–10 Days After Onset of ARS to Identify Culture-Proven Bacterial ARS in an Inception Cohort of 50 Military Recruits in Finland. N (%)†

Sensitivity (95% CI)

Specificity (95% CI)

LR1

LR2

Blockage

39 (78)

100 (63–100)

11 (3–27)

1.1

0

Clear discharge Purulent discharge

37 (74) 29 (58)

100 (63–100) 88 (47–98)

17 (7–34) 37 (21–55)

1.2 1.4

0 0.3

Symptom Present at 9–10 Days*

Nasal Symptoms

Postnasal discharge

27 (54)

63 (25–91)

37 (21–55)

1.0

1.0

29 (58)

88 (47–98)

37 (21–55)

1.4

0.3

Facial pain/pressure

16 (32)

50 (16–84)

66 (48–81)

1.5

0.8

Headache Otalgia

13 (26) 10 (20)

50 (16–84) 25 (4–65)

74 (57–87) 77 (60–90)

1.9 1.1

0.7 1.0

Toothache

2 (4)

13 (2–53)

97 (85–100)

4.4

0.9

Reduction of smell Pain Symptoms

Other Symptoms Sleeping disorder

16 (32)

50 (16–84)

66 (48–81)

1.5

0.8

Fatigue

29 (58)

63 (25–91)

31 (17–49)

0.9

1.2

Snoring Cough

6 (12) 38 (76)

25 (4–65) 88 (47–98)

89 (73–97) 11 (3–27)

2.2 1.0

0.8 1.1

*Interquartile range 9–10 days after onset of symptoms. † Symptom diaries were missing for 7 patients. ARS 5 acute rhinosinusitis; CI 5 confidence interval; LR1 5 positive likelihood ratio; LR– 5 negative likelihood ratio.

18–23). Of those, 48 (96%) of the participants were males, and 22 (44%) of them smoked. The median duration of acute symptoms was 2 days (IQR 2–3) at enrollment and 10 days (IQR 9–10) at the follow-up visit. Thirty-nine (78%) of the participants had respiratory virus nucleic acid in their nasopharynx at enrollment. The most frequent viruses were influenza A virus (n 5 20), adenovirus (n 5 20), and picornavirus group (n 5 21). At the follow-up visit, we performed a maxillary sinus puncture on 20 (40%) participants. Eight (16%) patients had positive cultures for nontypeable H. influenza and thus bacterial ARS. No other pathogenic bacteria were found in the aspiration sample cultures. The results were confirmed with bacterial PCR, as described earlier.16

LR1 of increasing toothache rose to 8.8, but again with only three patients.

Diagnostic Accuracy of Symptoms

DISCUSSION

At 9 to10 days, none of the symptoms proved to be ideal for diagnosing bacterial ARS (Table I). Absence of a relatively common symptom, patient-reported purulent nasal discharge, had a LR– of 0.3. Presence of a toothache was quite specific to bacterial ARS (LR1 4.4), but was a rare symptom with only two patients suffering from it.

Diagnostic Accuracy of Worsening of Symptoms A patient-reported increase in symptoms after 5 days changed the diagnostic accuracy of symptoms only slightly (Table II). Increasing purulent discharge had LR1 of 2.6, if present, and LR– of 0.4, if absent. The Laryngoscope 125: July 2015

Diagnostic Accuracy of Physical Examination Clinical findings at 9 to 10 days were much better than symptoms in predicting bacterial ARS (Table III). A finding of a moderate or profuse (versus none or minimal) amount of secretion in either nasal passage in anterior rhinoscopy had LR1 of 3.1 and LR– of 0.2. Any secretion seen in the posterior pharynx using a headlight and a tongue depressor involved LR1 of 5.3. If secretion was seen in the middle meatus using an endoscope, LR1 increased to 11.0. Furthermore, findings of both facial tenderness and cervical adenopathy moderately predicted a diagnosis of bacterial ARS (LR1 2.3 and 3.0, respectively).

We evaluated the usefulness of symptoms and physical findings in diagnosing culture-proven bacterial ARS in an inception cohort of young adults who had had acute respiratory symptoms, including nasal symptoms, for an average of 10 days. We found the presence of various patient-reported symptoms at 10 days to be a rather poor diagnostic test. Patient-reported worsening of symptoms (“double sickening”) did not improve diagnostic accuracy significantly. Toothache was the only symptom that accurately predicted bacterial ARS, but was present in only two patients. In contrast, several physical findings at 10 days were of more diagnostic use and frequent enough to be used for screening purposes. A finding of a moderate or profuse (vs. none or minimal) Autio et al.: Diagnosis of Bacterial Acute Rhinosinusitis

1543

TABLE II. Diagnostic Accuracy of the Worsening (“Double-Sickening”) of Various Symptoms After 5 Days From Onset of ARS to Identify Culture-Proven Bacterial ARS in an Inception Cohort of 50 Military Recruits in Finland. N (%)*

Sensitivity (95% CI)

Blockage

19 (38)

Clear discharge Purulent discharge

19 (38) 16 (32) 22 (44)

50 (16–84)

48 (31–66)

1.0

1.0

14 (28)

38 (9–75)

69 (51–83)

1.2

0.9

Worsening Symptom

Specificity (95% CI)

LR1

LR2

50 (16–84)

57 (39–74)

1.2

0.9

38 (9–75) 75 (35–96)

54 (37–71) 71 (54–85)

0.8 2.6

1.2 0.4

Nasal Symptoms

Postnasal discharge Reduction of smell Pain Symptoms Facial pain/pressure

14 (28)

50 (16–84)

71 (54–85)

1.8

0.7

Headache Otalgia

18 (36) 9 (18)

50 (16–84) 25 (4–65)

60 (42–76) 80 (63–92)

1.3 1.3

0.8 0.9

Toothache

3 (6)

25 (4–65)

97 (85–100)

8.8

0.8

Other Symptoms Sleeping disorders

12 (24)

38 (9–75)

74 (57–88)

1.5

0.8

Fatigue

20 (40)

63 (25–91)

57 (39–74)

1.5

0.7

Snoring Cough

5 (10) 20 (40)

13 (2–53) 38 (9–75)

89 (73–97) 51 (34–69)

1.1 0.8

1.0 1.2

Any Symptom

39 (78)

100 (63–100)

11 (3–27)

1.1

0

Any Nasal Symptom Any Pain Symptom

33 (66) 25 (50)

88 (47–98) 75 (35–96)

26 (13–43) 46 (29–63)

1.2 1.4

0.5 0.6

*Symptom diaries were missing for 7 patients. ARS 5 acute rhinosinusitis; CI 5 confidence interval; LR1 5 positive likelihood ratio; LR– 5 negative likelihood ratio.

amount of secretion in either nasal passage in the anterior rhinoscopy satisfactorily either ruled in, if present, or ruled out, if absent, bacterial ARS. If any secretion

was seen in the posterior pharynx, the probability of bacterial ARS increased markedly. This was true regardless of the quality of the secretion (clear or colored or

TABLE III. Diagnostic Accuracy of Various Physical Findings Recorded At 9–10 Days After Onset of ARS To Identify Culture-Proven Bacterial ARS in an Inception Cohort of 50 Military Recruits in Finland. Finding at 9–10 days*

N (%)

Sensitivity (95% CI)

Specificity (95% CI)

LR1

LR2

19 (38) 24 (48)

63 (25–91) 63 (25–91)

67 (51–80) 55 (39–70)

1.9 1.4

0.6 0.7

Clear Colored‡

14 (28) 27 (54)

13 (2–53) 88 (47–98)

69 (53–82) 52 (36–68)

0.4 1.8

1.3 0.2

Thick or crusty

25 (50)

75 (35–96)

55 (39–70)

1.7

0.5

19 (38)

88 (47–98)

71 (55–84)

3.1

0.2

8 (16) 6 (12)

50 (16–84) 50 (16–84)

91 (77–97) 95 (84–99)

5.3 11.0

0.6 0.5

10 (20) 11 (22)

38 (9–75) 50 (16–84)

83 (69–93) 83 (69–93)

2.3 3.0

0.8 0.6

Nasal Mucosa: Quality Erythematous Severe swelling† Nasal Secretion: Quality

Nasal Secretion: Quantity Moderate or profuse§ Nasal Secretion: Location Posterior pharynx# Nasal middle meatusk Other Findings Facial tenderness Cervical adenopathy

*Interquartile range 9–10 days after onset of symptoms † Inferior turbinate touching the nasal septum, Meltzer et al. 2006 (17) ‡ White, yellow, or green, Meltzer et al. 2006 (17) § Moderate or profuse (vs. none or minimal) amount of secretion, Meltzer et al. 2006 (17) # Any secretion seen in the posterior pharynx using a headlamp and a tongue depressor k Any secretion seen in the nasal middle meatus using a nasal endoscope ARS 5 acute rhinosinusitis; CI 5 confidence interval; LR1 5 positive likelihood ratio; LR– 5 negative likelihood ratio.

Laryngoscope 125: July 2015

1544

Autio et al.: Diagnosis of Bacterial Acute Rhinosinusitis

thick and crusty). A finding of cervical adenopathy additionally predicted bacterial ARS. Discovery of any secretion in the nasal middle meatus was the best predictor of bacterial ARS, but this entailed the use of an endoscope. The meaning of the present positive and negative LRs is clarified in the following clinical scenario. The prevalence of bacterial ARS in adult patients recruited in a general practice setting is estimated from a sinus puncture to be around 50%.9,18 Given that half of the symptomatic primary care patients suspected of having bacterial ARS actually have it results in a 50% pretest probability of bacterial ARS (probability scale 0%–100%). The best frequent enough symptom (i.e., a patientreported increase of purulent nasal discharge) resulted in a posttest probability of 29%, if absent, and 72%, if present. A physical finding of more than a minimal amount of secretion in either nasal passage after 10 days of symptoms gave a better posttest probability of 17%, if absent, and 76%, if present. If any secretion was seen in the pharynx, the posttest probability reached 84%, which even increased to 92% if secretion was seen in the nasal middle meatus. Our results emphasizing the importance of clinical findings in diagnosing bacterial ARS are in agreement with the meta-analysis of diagnostic tests for acute sinusitis by Varonen et al., if the magnitudes of the best clinical predictors’ LRs are considered.11 However, clinical examination was regarded as one unity in the systematic review; based on prior literature, no conclusion could be drawn on which components of the clinical examination were useful.9–11 Engels et al.9 and van den Broek et al.10 concluded in their systematic reviews that, based on the methodological flaws in earlier studies, no firm conclusion could be made on the usefulness of a clinical examination in making a diagnosis of bacterial ARS. Our results strongly contradict those of Hansen et al.,14 who concluded that a clinical examination is more or less worthless in diagnosis of bacterial ARS. Additionally, Hansen et al.14 found—similar to us—that a toothache accurately predicted bacterial ARS, but was too rare a symptom to be used as a screening test. To confirm the uniformity of the sample, we used an inception cohort design for which a group of patients near the onset of a rhinosinusitis episode was assembled. The symptoms and physical findings were collected blindly in a standardized way, and daily collection of symptoms enabled us to evaluate particularly the symptom progression pattern more accurately. We used the results of the sinus puncture and bacterial culture of the sinus fluid as our reference standard. The fact that we did not perform the puncture on all patients and all maxillary sinuses, but only on those sinuses that had CBCT findings indicative of bacterial sinusitis, may have involved information bias. However, CT of the paranasal sinuses is considered to be the imaging modality of choice when evaluating paranasal pathology.3 Our sample size was rather small, which is reflected in the relatively large CIs of the sensitivity and specificity values. However, we found several physical findings that predicted bacterial ARS more accurately than symptoms, Laryngoscope 125: July 2015

indicating that these signs are clinically significant. Still, it is possible that, because of the relatively small sample size and low proportion of cases who eventually got bacterial ARS, we did not identify all the variables predicting the presence of bacterial ARS. We performed our study among military recruits because we wanted to increase the proportion that would get bacterial ARS. Given the generally presented proportion of 2%3,4 who get bacterial sinusitis during ARS in primary care, we would have needed 400 such patients to get the eight cases of bacterial sinusitis we had in this sample. The military environment requires recognition concerning the generalizability of our results. We found H influenzae (NTHi) to be the only culture-proven pathogenic bacterium in the maxillary sinus, whereas others have reported S pneumonia and M catarrhalis to be common as well.19 The fact that our study was conducted among military recruits—among who NTHi is reported to be the most common nasal and paranasal pathogen20,21 — probably explains these differences. The diagnostic value of different symptoms and signs in bacterial ARS is probably not dependent on the type of the bacterium. The diverse viral etiology, as well as our earlier finding that viral etiology does not influence clinical signs and symptoms,16 further increase the generalizability of our findings. However, present findings warrant reconfirmation among other patients. Both U.S. and European guidelines3,4 also offer tools for separating viral and bacterial ARS, but they concentrate more on symptoms than clinical findings. Our results indicate that more accurate diagnostics could be achieved with a physical examination. The LRs of the best clinical items—namely findings of more than a minimal amount of secretion in either nasal passage and any secretion in the pharynx—are comparable with those of sinus radiography calculated in the systematic reviews by Varonen et al.11 and Engels et al.9 Moreover, evaluation of the quantity of nasal secretion in anterior rhinoscopy and the presence of pharyngeal secretion are simple enough tests to be performed by general practitioners.

CONCLUSION We found symptoms or their change to be of little use in identifying bacterial ARS. In contrast, we observed several clinical findings after 9 to 10 days of symptoms to predict bacterial ARS quite accurately. Therefore, we suggest careful examination of the patient and search for secretion in the nasal passages and posterior pharynx after 9 to 10 days of symptoms.

Acknowledgments We thank the Finnish Defence Forces for all the support in collecting the data, Soredex Oy (Finland) for providing the Scanora 3D imaging system (Soredex Inc) for CBCT, and Olympus Europa for the endoscopic equipment.

BIBLIOGRAPHY 1. Gill JM, Fleischut P, Haas S, Pellini B, Crawford A, Nash DB. Use of antibiotics for adult upper respiratory infections in outpatient settings: a national ambulatory network study. Fam Med 2006;38:349–354.

Autio et al.: Diagnosis of Bacterial Acute Rhinosinusitis

1545

2. Wang DY, Wardani RS, Singh K, et al. A survey on the management of acute rhinosinusitis among Asian physicians. Rhinology 2011;49:264–271. 3. Fokkens WJ, Lund VJ, Mullol J, et al. European position paper on rhinosinusitis and nasal polyps 2012. Rhinol Suppl 2012;23:3 p preceding table of contents, 1–298. 4. Chow AW, Benninger MS, Brook I, et al. IDSA clinical practice guideline for acute bacterial rhinosinusitis in children and adults. Clin Infect Dis 2012;54:e72–e112. 5. Bell BG, Schellevis F, Stobberingh E, Goossens H, Pringle M. A systematic review and meta-analysis of the effects of antibiotic consumption on antibiotic resistance. BMC Infect Dis 2014;14:13. doi: 10.1186/1471– 2334-14-13. 6. Panda S, El Khader I, Casellas F, et al. Short-term effect of antibiotics on human gut microbiota. PLoS One 2014;9:e95476. 7. Ahovuo-Saloranta A, Rautakorpi UM, Borisenko OV, Liira H, Williams JW Jr, Makela M. Antibiotics for acute maxillary sinusitis in adults. Cochrane Database Syst Rev 2014;2:CD000243. 8. Lemiengre MB, van Driel ML, Merenstein D, Young J, De Sutter AI. Antibiotics for clinically diagnosed acute rhinosinusitis in adults. Cochrane Database Syst Rev 2012;10:CD006089. 9. Engels EA, Terrin N, Barza M, Lau J. Meta-analysis of diagnostic tests for acute sinusitis. J Clin Epidemiol 2000;53:852–862. 10. van den Broek MF, Gudden C, Kluijfhout WP, et al. No evidence for distinguishing bacterial from viral acute rhinosinusitis using symptom duration and purulent rhinorrhea: a systematic review of the evidence base. Otolaryngol Head Neck Surg 2014;150:533–537. 11. Varonen H, Makela M, Savolainen S, Laara E, Hilden J. Comparison of ultrasound, radiography, and clinical examination in the diagnosis of

Laryngoscope 125: July 2015

1546

12. 13. 14. 15.

16.

17. 18.

19.

20.

21.

acute maxillary sinusitis: a systematic review. J Clin Epidemiol 2000; 53:940–948. Berg O, Carenfelt C. Etiological diagnosis in sinusitis: ultrasonography as clinical complement. Laryngoscope 1985;95:851–853. Berg O, Carenfelt C. Analysis of symptoms and clinical signs in the maxillary sinus empyema. Acta Otolaryngol 1988;105:343–349. Hansen JG, Schmidt H, Rosborg J, Lund E. Predicting acute maxillary sinusitis in a general practice population. BMJ 1995;311:233–236. Van Buchem L, Peeters M, Beaumont J, Knottnerus A. Acute maxillary sinusitis in general practice: the relation between clinical picture and objective findings. Eur J Gen Pract 1995;1:155–160. Autio TJ, Tapiainen T, Koskenkorva T, et al. The role of microbes in the pathogenesis of acute rhinosinusitis in young adults. Laryngoscope 2015;125:E1–E7. Meltzer EO, Hamilos DL, Hadley JA, et al. Rhinosinusitis: developing guidance for clinical trials. J Allergy Clin Immunol 2006;118:S17–S61. Smith SS, Ference EH, Evans CT, Tan BK, Kern RC, Chandra RK. The prevalence of bacterial infection in acute rhinosinusitis: a systematic review and meta-analysis. Laryngoscope 2015;125:57–69. Payne SC, Benninger MS. Staphylococcus aureus is a major pathogen in acute bacterial rhinosinusitis: a meta-analysis. Clin Infect Dis 2007;45: e121–e127. Jousimies-Somer HR, Savolainen S, Ylikoski JS. Bacteriological findings of acute maxillary sinusitis in young adults. J Clin Microbiol 1988;26: 1919–1925. Jousimies-Somer HR, Savolainen S, Ylikoski JS. Comparison of the nasal bacterial floras in two groups of healthy subjects and in patients with acute maxillary sinusitis. J Clin Microbiol 1989;27:2736–2743.

Autio et al.: Diagnosis of Bacterial Acute Rhinosinusitis

Diagnostic accuracy of history and physical examination in bacterial acute rhinosinusitis.

To evaluate the diagnostic accuracy of symptoms, the symptom progression pattern, and clinical signs in identifying bacterial acute rhinosinusitis (AR...
98KB Sizes 0 Downloads 6 Views