2248
SJS102210.1177/1457496913482248Hyperbilirubinemia as a predictor for appendiceal perforation: a systematic reviewJ. Burcharth, et al.
Scandinavian Journal of Surgery 102: 55–60, 2013
HYPERBILIRUBINEMIA AS A PREDICTOR FOR APPENDICEAL PERFORATION: A SYSTEMATIC REVIEW
J. Burcharth, H. C. Pommergaard, J. Rosenberg, I. Gögenur Department of Surgery D, Herlev Hospital, University of Copenhagen, Herlev, Denmark ABSTRACT
Background and Aims: Delayed or wrong diagnosis in patients with appendicitis can result in perforation and consequently increased morbidity and mortality. Serum bilirubin may be a useful marker for appendiceal perforation. The purpose of this systematic review was to evaluate studies investigating elevated serum bilirubin as a predictor for appendiceal perforation. Material and Methods: Medline, Embase, and Cochrane databases were searched for studies evaluating elevated bilirubin in the diagnosis of perforated appendicitis. Study selection criteria included English language papers evaluating serum bilirubin as a marker of appendiceal perforation in humans. A total of 189 abstracts were screened for eligibility, of which five clinical studies were included in this study. Results: Bilirubin was significantly higher in patients with appendiceal perforation compared with patients with appendicitis without perforation. Elevated serum bilirubin had a sensitivity ranging from 0.38 to 0.77 and a specificity ranging from 0.70 to 0.87 in predicting appendiceal perforation. Conclusions: Elevated serum bilirubin for determining the risk of perforation in appendicitis has low sensitivity but higher specificity. This measure can therefore be used as a supplement in the diagnostic process. Key words: Surgery, appendix, appendicitis, perforation, hyperbilirubinemia, bilirubin, serological markers BACKGROUND Acute appendicitis is one of the most frequent surgical conditions, and the lifetime risk for appendectomy is estimated to be 12% for males and 23% for females (1). Acute uncomplicated appendicitis can be difficult to distinguish clinically from perforated appendicitis, especially in the elderly and in children (2, 3). Simple acute appendicitis can be a minor Correspondence: Jakob Burcharth, M.D. Department of Surgery D Herlev Hospital University of Copenhagen Herlev Ringvej 75 DK-2730 Herlev Denmark Email:
[email protected]
surgical procedure with limited complications; however, delayed diagnosis and subsequently appendiceal perforation can result in a potentially lethal condition (4). The mortality associated with simple acute appendicitis is reported to be 0.3% but increases to 6% in cases with perforation (5). Early diagnosis of appendiceal perforation is important to limit the associated abdominal sepsis. Moreover, where radiological modalities such as computed tomography (CT) scan and ultrasound (US) are effective in supplementing the diagnosis of acute appendicitis (6, 7), both modalities have lower sensitivity in detecting perforated appendicitis (8, 9). Development of supplementary tools besides clinical examination and radiology could be beneficial in the early diagnosis (5). Several studies have found bilirubin to be a useful serological marker for predicting acute appendicitis (10, 11) and appendiceal perforation (5, 11–14).
56
J. Burcharth, et al.
The purpose of this systematic review was to evaluate the current literature regarding elevated serum bilirubin as a diagnostic marker for appendiceal perforation. METHODS LITERATURE SEARCH
Using the preferred reporting items for systematic reviews and meta-analysis (PRISMA) (15, 16) guidelines, a systematic literature search was performed in Medline (January 1966–February 2012), Embase (January 1980–February 2012), and Cochrane databases (April 1996 to February 2012). The aim of the search strategy was to identify human clinical studies evaluating serum bilirubin as a prognostic factor for appendiceal perforation. In Medline, the following search terms were used: ((appendix(MeSH Terms) OR appendix (All Fields)) OR (appendicitis (MeSH Terms) OR appendicitis (All Fields))) AND ((hyperbilirubinaemia (All Fields) OR hyperbilirubinaemia (MeSH Terms) OR hyperbilirubinemia (All Fields)) OR (hyperbilirubinemia (MeSH Terms) OR bilirubinemia (All Fields)) OR (jaundice (MeSH Terms) OR jaundice (All Fields)) OR (bilirubin (MeSH Terms) OR bilirubin (All Fields))). The Embase and Cochrane search used the multifield advanced search function with the same search terms. STUDY SELECTION CRITERIA
Regarding language, only English language studies were included. Only original human studies on admitted patients with perforated appendicitis (histologically or surgically verified) and evaluating hyperbilirubinemia were included. The primary study outcome was evaluation of serum bilirubin as a diagnostic marker in acute perforated appendicitis compared with simple acute appendicitis without perforation. The outcome was measured either by absolute numeric values or by specific cutoff values and estimates of sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV). Studies not evaluating bilirubin values and studies evaluating the use of elevated bilirubin for simple acute appendicitis compared with no appendicitis were not included. For details on specific study selection, see Fig. 1. INITIAL ABSTRACT SCREENING AND FULL TEXT ASSESSMENT
Titles and abstracts were independently assessed for eligibility by two reviewers according to the selection criteria. Any uncertainties between the reviewers were settled by discussion. Full manuscripts were obtained and evaluated in detail from the studies assessed suitable for inclusion. DATA EXTRACTION
Data extraction was performed on five studies. The following data were extracted: study aims, study design, patient material (number, subgroups, demographics), primary and secondary outcomes, bilirubin values, and conclusions. Data were extracted by both
reviewers. Data were extracted directly from the studies; however, conversion of the volume of serum bilirubin into micromole per liter was performed in order to standardize the data (Table 1). RESULTS A total of 265 possible studies were identified, and after doublets were removed, the search strategy resulted in 189 possible studies for inclusion (Fig. 1). A total of 2243 patients were included in the five studies (Table 1). All studies were retrospective and descriptive in design. No healthy control group was used in any of the studies. All patients included were operated for appendicitis, which was verified by pathological examination; however, in two studies, a subgroup of the patients had pathologically normal appendices (5, 11). The levels of bilirubin were compared between patients with perforated appendicitis and patients with simple appendicitis without perforation. There were significantly higher levels of serum bilirubin among patients with perforated appendicitis compared with patients without perforation with p values ranging from p = 0.01 to p < 0.001 (Table 1). In the included studies, serum bilirubin was measured as values either above certain threshold values (>17.1 or >20 µmol/L) or in absolute values. The mean serum bilirubin values for patients with perforated appendicitis ranged from 23.1 (11) to 25.7 µmol/L (5). The sensitivity for elevated bilirubin varied among the studies with threshold values. The sensitivity varied between 0.38 and 0.77 and the specificity varied from 0.70 to 0.87. NPVs were between 82% and 93%, and PPVs were between 21% and 51%. DISCUSSION Serum bilirubin was significantly elevated in patients with perforated appendicitis compared with patients with simple acute appendicitis without perforation. In general, the patients with perforated appendicitis had mean serum bilirubin values ranging from 23 to 26 µmol/L (Table 1). Depending on the specific threshold values, the specificity was generally high. Moreover, NPV was high for elevated serum bilirubin, but PPV was generally low. In the studies included in this review, C-reactive protein (CRP) and white blood cell (WBC) count have been evaluated as predictors for appendiceal perforation. It seems that CRP generally has a higher sensitivity but lower specificity compared with bilirubin (5, 11, 14). However, the sensitivity and specificity of CRP and bilirubin are determined by the threshold values. The optimal condition is a maximum value of sensitivity combined with a maximum value of specificity. Therefore, a threshold value has to be calculated from the optimal sensitivity and specificity, which is normally done with the aid of a receiver operating characteristics (ROC) curve. In general, when setting a low threshold value, this generates a high sensitivity and low specificity and vice versa when setting a high threshold. Both CRP and bilirubin have to be measured in absolute values, and in this situation, bilirubin
Hyperbilirubinemia as a predictor for appendiceal perforation: a systematic review
57
Possible relevant manuscripts in the literature search (n = 265) Databases: - PubMed (n = 93) - Embase (n = 172) - Cochrane (n = 0) Limits: Only English language manuscripts were accepted Only original (prospecve / retrospecve) arcles examining elevated blood bilirubin were accepted
Abstracts screened aer duplicates removed from combined searches (n = 189)
Full-text manuscripts read and assessed for eligibility (n = 9)
Studies excluded (n = 180) - Title or abstract revealed not appropriate (n = 166) - Non-Englishlanguage manuscripts (n = 3) - Case reports (n = 6) - Leer to the editor (n = 2) - Not available (n = 3)
Manuscripts excluded (inclusion/exclusion criteria) (n = 4) - Case report (n = 2) - Not evaluaon bilirubin (n = 1) - Only evaluang non-perforated appendicis (n = 1)
Included manuscripts in systemac review (n = 5)
Fig. 1. Flowchart of literature search and included studies.
seems superior in predicting perforated appendicitis. However, even though both CRP and bilirubin are mentioned in the included studies, none of the studies presented data on the combined use of CRP and bilirubin as predictors of appendiceal perforation. Diagnostic scores have been developed in diagnosing acute appendicitis such as the Alvarado score (scale 0–10) (17), the modified Alvarado score for use in pediatric patients (18), pediatric appendicitis score (PAS; scale 0–10) (3), RajaIsteri Pengiran Anak Saleha Appendicitis (RIPASA) score for use in Asian patients (scale 0–14) (19), and appendicitis inflammatory response score (AIRS; scale 0–12) (20). Common for these scoring systems is the use of symptoms (duration of pain, migration of pain, nausea, vomiting), signs (tenderness, fever), and investigations (leukocytosis, CRP). All of these scores have been proven useful in predicting acute appendicitis in patients presenting
with pain in the lower right fossa, but none of them evaluate the risk of appendiceal perforation. Furthermore, none of the above-mentioned scores use elevated bilirubin as a marker. It has been shown that during appendicitis, an ulceration of the mucosa in the appendix occurs due to the inflammation, which facilitates bacterial translocation from appendix to the portal blood system (21). The most common bacteria to infect the appendix is the gram-negative Escherichia coli. When E. coli reaches the hepatic tissue via the portal venous system, animal models have shown that the bacteria interferes with the hepatocyte microcirculation, which induces damage to the liver cells and compromises excretion of bile acids into the bile canaliculi (22). Furthermore, E. coli has been shown to induce intravascular hemolysis, and both mechanisms may result in an increased amount of bilirubin circulating in the
Group 1: acute suppurative appendicitis (n = 116). Age (median (range)): 33 (5–66) years Group 2: gangrenous/perforated appendicitis (n = 41). Age (median (range)): 31 (7–61) years
n = 55 femalesa n = 102 malesa
n = 306 females n = 232 males
NR
n = 145 females n = 206 males
n = 218 females n = 254 males
Estrada et al. (13)
Sand et al. (5)
Käser et al. (14)
Atahan et al. (12)
Emmanuel et al. (11)
Bilirubin threshold >20.5 µmol/L Group 2: 18.7 µmol/L Group 4: 23.1 µmol/L (p = 0.01)
Bilirubin threshold: >20 µmol/L Group 1: 22.3% Group 2: 38% (p < 0.001) Bilirubin threshold: >17.1 µmol/L (mean ± SD) Group 1: 9.1 ± 0.3 Group 2: 11.1 ± 0.5 Group 3: 21.9 ± 0.4 (p = 0.001)
Bilirubin threshold: >17.1 µmol/L Group 1: 31% elevated bilirubin Group 2: 56% elevated bilirubin (p = 0.004) Bilirubin threshold: >17.1 µmol/L (mean ± SD) All groups combined: 15.4 µmol/L ± 0.6 Group 4: 25.7 µmol/L ± 0.9 (p < 0.05)
Results
OR: odds ratio; PPV: positive predictive value; NPV: negative predictive value; NR: not registered. aNumber of males or females calculated on the basis of percentage numbers.
Group 1: no pathologic findings (n = 67). Age: 27 years (53 females, 14 males). Group 2: reactive follicular hyperplasia (n = 95). Age: 32.5 years (64 females, 31 males) Group 3: acute appendicitis (n = 263). Age: 49.2 years (145 females, 118 males). Group 4: perforated acute appendicitis (n = 97). Age: 48.3 years (38 females, 59 males) Group 5: Acute necrotizing appendicitis (n = 16). Age: 35.7 years (6 females, 10 males) Group 1: no perforation (n = 570). Age (range): 4.9–86.9 years Group 2: perforation (n = 155). Age (range): 9.7–92.4 years Age: >16 years Group 1: acute appendicitis or reactive hyperplasia (n = 49) (32 females, 17 males). Group 2: phlegmonous appendicitis (n = 257) (98 females, 159 males) Group 3: perforated appendicitis (n = 45) (15 females, 30 males) Age (mean, range): 27 (5–82) years Group 1: no inflammation (n = 86) Group 2: acute appendicitis (n = 294) Group 3: acute appendicitis (transmural inflammation) (n = 47) Group 4: perforation/gangrenous appendicitis (n = 45)
Design
Study patient material
TABLE 1. Studies included in the systematic review.
60%/70%
77%/87%
38%/78%
70%/86%
NR/NR
Sensitivity/ specificity
21%/92%
45%/97%
32%/82%
51%/93%
NR/NR
PPV/NPV
Elevated bilirubin level is predictor for acute appendicitis and perforation
Both bilirubin and CRP are usable predictors of perforation; however, CRP is superior Bilirubin together with infectious markers can be used alone and is an early predictor of perforation
Elevated bilirubin together with clinical symptoms can predict appendiceal perforation
Bilirubin can be used as a predictor of perforation. OR 2.96 for perforation if elevated bilirubin
Conclusion
58 J. Burcharth, et al.
Hyperbilirubinemia as a predictor for appendiceal perforation: a systematic review
blood (23). Moreover, elevated serum bilirubin can be a result of mechanisms other than hepatobiliary diseases or intravascular hemolysis (24). Bilirubin can be elevated in cases of sepsis, intra-abdominal abscesses from urological, gynecological or gastroenterological origins, antiviral therapy or in patients with genetic diseases such as Dubin–Johnson syndrome, Rotor’s syndrome, and Gilbert’s syndrome (25). Dubin– Johnson and Rotor’s syndrome are uncommon, whereas Gilbert’s syndrome is a benign disease where unconjugated serum bilirubin is slightly elevated, and it is estimated that 3%–10% of Caucasians have this condition (26). Gilbert’s syndrome is remarkably more prevalent among males than females (27). It has been proposed that hyperbilirubinemia is a weaker marker of appendiceal perforation among persons with Gilbert’s syndrome (14). However, since Gilbert’s syndrome is considered a rare condition, this may not affect the value of hyperbilirubinemia as an overall predictor in both males and females. There exists no single diagnostic test or symptom other than surgery with pathologic examination that can definitely result in a diagnosis. Several prognostic factors such as duration of symptoms, pain migration, indirect pain, abdominal guarding, and fever can be used in combination to predict a possible perforation of the appendix (28, 29). The studies included in this systematic review have all shown that elevated bilirubin can be used as a prognostic factor for the assessment of perforated appendicitis. The above-mentioned combinations of factors are especially important in the evaluation of patients with atypical clinical presentation. Furthermore, in many departments, blood samples including liver functional tests are not standard in children, and often only clinical evaluation is used for diagnosis. Regarding the clinical use of predictors in diagnostics of appendicitis or appendiceal perforation, elevated bilirubin should be considered as a supplement to other diagnostic tools such as CT scans. Data show that CT findings predictive of perforated appendicitis such as presence of extraluminal gas and localized abscess have high specificity of more than 98%, however, with lower sensitivity of 40% (9). A recent study found that the use of a low-dose abdominal CT scan in diagnosing simple appendicitis resulted in a sensitivity and specificity both exceeding 93% (30). However, early perforation or microperforation of the appendix can be very difficult to diagnose using CT (9). The included studies had several limitations. They were all retrospective and without a healthy control group. The largest study (14) included 725 patients from two different centers, and the study period was different among these centers (5 years in one center and 10 years in the other center), which might have made treatment and diagnostic regimens not comparable in the study. Moreover, possible detection bias was present since the study used different measurement methods to evaluate bilirubin values. These methods were only comparable at the cutoff value and not at the high and low extremes, which makes high and low values of bilirubin not comparable within the study (14). Moreover, a possible selection
59
bias could have been introduced since bilirubin was not measured in all patients. This was also the case in the other included studies (11, 13). The gender distribution was significantly different between the study groups in one study (12) and in two other included studies; it was not reported whether the gender difference between the study groups was statistically significant (5, 14). An uneven distribution of males and females could result in a risk of Gilbert’s syndrome being overrepresented in one group, which would distort the results. Moreover, possible confounders are present in all studies. This confounding can occur when the surgeon is aware that elevated bilirubin can predict perforation. Thus, the surgeon might be more prone to operate in case of an elevated bilirubin and therefore be more likely to diagnose appendicitis, which under different circumstances might not have been diagnosed, for example, subclinical appendicitis that may resolve without treatment. In conclusion, elevated serum bilirubin can be used as a supplemental diagnostic tool in perforated appendicitis. The other important supplemental diagnostic tool for appendicitis and appendiceal perforation is CT scan that has high specificity for appendiceal perforation, although clinically, it is limited by low sensitivity. It is important to stress that no single predictive factor (serological, radiological) can stand alone in determining the course of action in patients suspected of appendicitis. Clinical examination remains the cornerstone in the diagnosis of simple and perforated appendicitis. Further studies should evaluate whether the predictive values of bilirubin, CRP, and WBC combined would be a stronger diagnostic tool for the clinician. REFERENCES 1. Addiss DG, Shaffer N, Fowler BS et al: The epidemiology of appendicitis and appendectomy in the United States. Am J Epidemiol 1990;132:910–925. 2. Konan A, Hayran M, Kilic YA et al: Scoring systems in the diagnosis of acute appendicitis in the elderly. Ulus Travma Acil Cerrahi Derg 2011;17:396–400. 3. Escribá A, Gamell AM, Fernández Y et al: Prospective validation of two systems of classification for the diagnosis of acute appendicitis. Pediatr Emerg Care 2011;27:165–169. 4. Ditillo MF, Dziura JD, Rabinovici R: Is it safe to delay appendectomy in adults with acute appendicitis? Ann Surg 2006;244:656–660. 5. Sand M, Bechara FG, Holland-Letz T et al: Diagnostic value of hyperbilirubinemia as a predictive factor for appendiceal perforation in acute appendicitis. Am J Surg 2009;198:193–198. 6. Jo YH, Kim K, Rhee JE et al: The accuracy of emergency medicine and surgical residents in the diagnosis of acute appendicitis. Am J Emerg Med 2010;28:766–770. 7. Doria AS: Optimizing the role of imaging in appendicitis. Pediatr Radiol 2009;39(Suppl. 2):S144–S148. 8. Rybkin AV, Thoeni RF: Current concepts in imaging of appendicitis. Radiol Clin North Am 2007;45:411–422. 9. Bixby SD, Lucey BC, Soto JA et al: Perforated versus nonperforated acute appendicitis: Accuracy of multidetector CT detection. Radiology 2006;241:780–786. 10. McGowan DR, Sims HM, Shaikh I et al: The value of hyperbilirubinaemia in the diagnosis of acute appendicitis. Ann R Coll Surg Engl 2011;93:498.
60
J. Burcharth, et al.
11. Emmanuel A, Murchan P, Wilson I et al: The value of hyperbilirubinaemia in the diagnosis of acute appendicitis. Ann R Coll Surg Engl 2011;93:213–217. 12. Atahan K, Üreyen O, Aslan E et al: Preoperative diagnostic role of hyperbilirubinaemia as a marker of appendix perforation. J Int Med Res 2011;39:609–618. 13. Estrada JJ, Petrosyan M, Barnhart J et al: Hyperbilirubinemia in appendicitis: A new predictor of perforation. J Gastrointest Surg 2007;11:714–718. 14. Käser SA, Fankhauser G, Willi N et al: C-reactive protein is superior to bilirubin for anticipation of perforation in acute appendicitis. Scand J Gastroenterol 2010;45:885–892. 15. Moher D, Liberati A, Tetzlaff J et al: Preferred reporting items for systematic reviews and meta-analyses: The PRISMA statement. PLoS Med 2009;6:e1000097. 16. Liberati A, Altman DG, Tetzlaff J et al: The PRISMA statement for reporting systematic reviews and meta-analyses of studies that evaluate health care interventions: Explanation and elaboration. J Clin Epidemiol 2009;62:e1–e34. 17. Alvarado A: A practical score for the early diagnosis of acute appendicitis. Ann Emerg Med 1986;15:557–564. 18. Kalan M, Talbot D, Cunliffe WJ et al: Evaluation of the modified Alvarado score in the diagnosis of acute appendicitis: A prospective study. Ann R Coll Surg Engl 1994;76: 418–419. 19. Chong CF, Thien A, Mackie AJ et al: Comparison of RIPASA and Alvarado scores for the diagnosis of acute appendicitis. Singapore Med J 2011;52:340–345. 20. Andersson M, Andersson RE: The appendicitis inflammatory response score: A tool for the diagnosis of acute appendicitis that outperforms the Alvarado score. World J Surg 2008;32:1843–1849.
21. Sisson RG, Ahlvin RC, Harlow MC: Superficial mucosal ulceration and the pathogenesis of acute appendicitis. Am J Surg 1971;122:378–380. 22. Rink RD, Kaelin CR, Giammara B et al: Effects of live Escherichia coli and Bacteroides fragilis on metabolism and hepatic pO2. Circ Shock 1981;8:601–611. 23. Shander A: Anemia in the critically ill. Crit Care Clin 2004;20:159–178. 24. Franson TR, Hierholzer WJ, LaBrecque DR: Frequency and characteristics of hyperbilirubinemia associated with bacteremia. Rev Infect Dis 2012;7:1–9. 25. Buyukasik Y, Akman U, Buyukasik NS et al: Evidence for higher red blood cell mass in persons with unconjugated hyperbilirubinemia and Gilbert’s syndrome. Am J Med Sci 2008;335:115–119. 26. Bailey A, Robinson D, Dawson AM: Does Gilbert’s disease exist? Lancet 1977;1:931–933. 27. Berg C, Crawford J, Gollan J: Bilirubin metabolism and the pathophysiology of jaundice. In: Schiff E, Sorrell M, Maddrey W (Eds) Schiff’s diseases of the liver. Lippincott-Raven, Philadelphia, PA, 1999, pp. 147–192. 28. Andersson REB: Meta-analysis of the clinical and laboratory diagnosis of appendicitis. Br J Surg 2004;91:28–37. 29. Oliak D, Yamini D, Udani VM et al: Can perforated appendicitis be diagnosed preoperatively based on admission factors? J Gastrointest Surg 2000;4:470–474. 30. Kim K, Kim S, Kim SY et al: Low-dose abdominal CT for evaluating suspected appendicitis. N Engl J Med 2012;366: 1596–1605.
Received: May 9, 2012 Accepted: September 11, 2012
