Dominik A. Walczak 1,2, Dariusz Pawełczak 2, Agata Żółtaszek 3, Karolina Ptasińska 1, Piotr W. Trzeciak 1, Zbigniew Pasieka 2

Transcription

1 POLSKI PRZEGLĄD CHIRURGICZNY 2015, 87, 2, /pjs The value of scoring systems for the diagsis of acute appendicitis Dominik A. Walczak 1,2, Dariusz Pawełczak 2, Agata Żółtaszek 3, Rajmund Jaguścik 1, Wojciech Fałek 1, Monika Czerwińska 1, Karolina Ptasińska 1, Piotr W. Trzeciak 1, Zbigniew Pasieka 2 Department of General Surgery, John Paul II Memorial Hospital in Bełchatów 1 Ordynator: lek. P. W. Trzeciak Department of Experimental Surgery, Medical University in Łódź 2 Kierownik: dr hab. Z. Pasieka, prof. nadzw. Department of Spatial Ecometrics, Faculty of Ecomics and Sociology, University in Łódź 3 Kierownik: prof. dr hab. J. Suchecka Accurate diagsis of acute appendicitis (AA) is still a problem and is t always easy, even for experienced surgeons. Studies have shown that 20 to 30% of the appendices removedwere rmal. Therefore, various scoring systems have been developed to aid in the diagsis of doubtful cases and reduce the number of unnecessary appendectomies. The aim of this study was to assess the diagstic value of different scoring systems in acute appendicitis. Material and methods. The study involved 94 patients who underwent laparotomy due to suspected acute appendicitis. Medical examination at hospital admission was performed by a resident and a general surgery specialist. The probability of AA was evaluated using six different scoring systems: Alvarado, Fenyo, Eskelinen, Ohman, Tzankis, and RIPASA. The resident calculated the results in individual systems. The decision to perform the operation was taken by a specialist surgeon who did t kw the results. Results. Normal appendix was removed in 26% of cases. Eskelinen, RIPASA and Alvarado systems showed highest sensitivity (99%, 88% and 85%, respectively). Tzankis and Fenyo systems showed highest specificity (62% and 50%, respectively). Conclusions. Our study has shown limited value of scoring systems for the diagsis of acute appendicitis.the systems may improve diagstic accuracy as they require obtaining a more detailed medical history, and making a more thorough and organized data analysis. However, the scoring systems should be treated only as an aid to diagsis. Key words: acute appendicitis, scoring systems, diagsis, appendectomy, symptoms, signs Acute appendicitis (AA) is the most frequent cause of peritonitis in patients hospitalized at general surgery departments. Despite growing accuracy of diagstic methods, the level of diagstic errors has remained around 20-30% for many years (1). Additionally, in women aged years the percentage of unnecessary laparotomies may reach even 45.6% (2). Certainly, in some cases, even though the initial diagsis of AA proves to be wrong, the operation turns out to be justified when other cause of the complaint is found, e.g. caecal tumour, twisted or ruptured ovarian cyst, or extrauterine pregnancy. The core of the problem are the cases when the operation fails to indicate the cause of the pain, which, in all likelihood, is of a n-surgical nature. In such cases, the macroscopically rmal appendix is often removed, but the procedure may be considered unnecessary (3). Undoubtedly, over the 127 years since the first successful appendectomy, mortality and number of complications have decreased significantly. However, this does t mean that

2 66 D. A. Walczak et al. there are complications at all. Sometimes the complications result in death of a young and healthy person. For example, in a large Swedish population-based study, 1 death per every 5,000 appendectomies in patients aged up to 50 years was reported (4). Additionally, a patient is exposed to the risk of wound infection, development of hernia, or mechanical ileus, usually caused by adhesions after appendectomy. The complications take on a whole new dimension when eventually the appendix turns out t to be the cause of the complaints, so the removal was unjustified (3). On the other hand, it is worth mentioning that the errors may result in delaying accurate diagsis, leading to appendix perforation, then diffuse peritonitis, and consequently to related complications. The use of scoring systems in AA diagsis is t a new concept. More than ten such systems have been developed since the beginning of the 1980s to present (5-11). The systems assign a point value to a symptom obtained from the patient s medical history, physical examination or additional examinations, to eventually determine the probability of AA in the patient. They are meant to provide more thorough and organized data analysis, and thus to improve diagstic accuracy. The aim of our study was to assess the value of different scoring systems for the diagsis of acute appendicitis. Material and methods Between January 2010 and December 2012, 156 patients with suspected acute appendicitis were hospitalized in the General Surgery Department. Each patient was ex- Table 1. Alvarado scoring system Symptom/sign Migration of pain 1 Arexia 1 Nausea/vomiting 1 Right iliac fossa tenderness 2 Fever 1 Positive Blumberg sign 1 Leukocytosis >10,000 2 Left shift in WBC differential 1 1-4: low probability of AA; 5-6: average probability, observation recommended; 7-10: high probability of AA amined at admission by a general surgery specialist and a resident. The total of 10 specialists and 4 residents took part in the study. All patients underwent basic laboratory tests, including complete blood count, c-reactive protein level, general urinalysis, and abdominal ultrasound performed by a specialist radiologist. The resident filled the prepared questionnaire regarding the diagsed symptoms and tests results, then calculated the probability of acute appendicitis in six different scoring systems: Alvarado (the assumed cut-off point (COP) for the diagsis of AA: 7 points; tab. 1), Fenyo (COP: -2 points; tab. 2), Eskelinen (COP: 55 points; tab. 3), Ohman (COP: 12 points; tab. 4), Tzankis (COP: 8 points; tab. 5), and RIPASA (COP: 7.5 points; tab. 6). The decision to perform the operation was taken by a general surgery specialist who did t kw the results in the individual systems. Eventually 92 patients were subjected to laparotomy, and this group was Table 2. Fenyo scoring system Symptom/sign, variable All patients start with -10 Sex: female -15 male +8 WBC count < <14000 Duration of complaint: <24 h h >48 h Progression of pain Migration of pain Vomiting Aggravation with cough Positive Blumberg sign Rigidity -2 or more: AA, -17 or less: n-specific pain

3 The value of scoring systems for the diagsis of acute appendicitis 67 Table 3. Eskelinen scoring system Symptom/sign Point criterion Ratio Tenderness, pain 2 right iliac fossa 11,41 1 any other site Rigidity 2 6,62 1 WBC count 2 > ,88 1 < Positive Blumberg 2 4,25 sign 1 Pain at examination 2 right iliac fossa 3,51 1 any other site Duration of complaint 2 <48 h 1 >48 h 2,13 AA 55 points Table 4. Ohman scoring system Symptom/sign, variable Right iliac fossa pain 4,5 Positive Blumberg sign 2,5 No dysuric symptoms 2 Continuous pain 2 WBC count >10,000 1,5 Age < 50 years 1,5 Migration of pain to the right lower 1 quadrant Muscular tone 1 <6: low probability of AA; : average probability, observation recommended; >11.5: high probability of AA qualified for further analysis. After the operation, the surgeon filled in a questionnaire regarding the course of the operation, condition of the appendix, and other diagsed pathologies that may have caused the pain. Finally, the diagsis of acute appendicitis was verified by postoperative histopathological examination. The statistical analysis was performed using the Pearson independence test and ANOVA variance analysis. The adopted level of significance was p=0.05. Results The analysed group of 92 patients who underwent laparotomy due to suspected acute appendicitis included 46 women and 46 men. The mean patient age was 38 years (range: years). Normal appendix was removed in 24 patients (26% of cases). Eleven patients were intraoperatively diagsed with other pathology which caused their complaints. It Table 5. Tzakis scoring system Symptom/sign, variable Signs of appendicitis in ultrasound 6 examination Right iliac fossa pain 4 Positive Blumberg sign 3 WBC count >12,000 2 AA 8 points Table 6. RIPASA scoring system Male 0,5 Female 1 Age <39.9 years 1 Age >40 years 0,5 Right iliac fossa pain 0,5 Migration of pain 0,5 Arexia 1 Nausea/vomiting 1 Complaint duration <48 h 1 Duration of complaint >48 h 0,5 Right iliac fossa tenderness 1 Obrona mięśniowa / muscular guarding 2 Positive Blumberg sign 1 Positive Rovsing sign 2.0 Fever >37 o C, <39 o C 1 Leukocytosis 1 Negative urinalysis 1 Foreigner 1 <5: AA very unlikely; 5-7.0: AA unlikely; : AA very likely; >12: definitely AA was mesenteric lymphadenitis in 6 patients, ruptured ovarian cyst in 2 patients, extrauterine pregnancy in 1 patient, acute pancreatitis in 1 patient, and necrotic omental fragment in 1 patient. In the subgroup of female patients up to 40 years old (30 subjects), rmal appendix was removed in 33% of cases (10 patients), 23% of which (7 patients) were t diagsed with other surgical cause of the complaint. None of the studied symptoms and tests results was significantly correlated with AA diagsis (tab. 7). The highest score was obtained for: positive Rovsing sign (p=0.06), arexia (p=0.1), fever (p=0.11) and suspected AA in ultrasound examination (p=0.15). Results of ultrasound diagsis were thoroughly analysed. In our study, the sensitivity of ultrasound was 38% and its specificity 83%. The accuracy of the diagsis depended largely on the radiologist s experience, and thus the sensitivity for individual physicians ranged from 75% to 20%.

4 68 D. A. Walczak et al. Table 7. Statistical significance of individual symptoms Symptom/sign p Symptom/sign p Sex 0,56 Cough 0,93 Fever >37 o C, <39 o C 0,11 Muscular Guardini 0,96 Nausea 0,47 Nature of pain: continuous/intermittent 0,9 Arexia 0,1 Dysuric symptoms 0,54 Duration of complaint 0,23 AA evidences in ultrasound examination 0,15 Pain at palpation 0,54 Deviations in urinalysis 0,55 Migration of pain 0,56 WBC count 0,17 Positive Blumberg sign 0,82 CRP level 0,93 Positive Rovsing sign 0,06 Table 8. Results for individual scoring systems Scoring Scale Sensitivity Specificity PPV NPV p Alvarado 85% 16% 74% 29% 0,9 Fenyo 44% 50% 71% 25% 0,79 Eskelinen 99% 8% 75% 50% 0,45 Tzakis 65% 62% 81% 40% 0,13 RIPASA 88% 9% 68% 20% 0,72 Results for individual scoring systems is presented in tab. 8. Eskelinen scoring system showed the highest sensitivity, and Tzakis system showed the highest specificity. None of the progstication systems is significantly correlated with AA diagsis. Discussion The last decades have seen an increase in the probabilistic approach in clinical decisionmaking. This has resulted in the development of a number of scoring systems, computer models and algorithms. Similarly, many systems have been developed for the diagsis of acute appendicitis, which, according to their authors, are intended to provide an additional tool to facilitate accurate diagsis and management of AA. This is important, insomuch as the level of diagstic errors has remained the same for years despite growing accuracy of diagstic methods. The new AA markers, such as interleukin-6, lactoferrin, calprotectin, serum amyloid A, myeloperoxidase and many other, proved to be either of diagstic value, or impractical despite positive results of pilot studies, for instance due to poor availability (12, 13, 14). Many publications emphasize high value of computed tomography (CT) in diagsing AA, particularly in doubtful cases. This examination allows to reduce the percentage of unnecessary appendectomies, on average to several percent (15-18). In an American study, half of the patients, whose CT scan did t confirm acute appendicitis, were diagsed with other real cause of the complaint (18). Unfortunately, Polish conditions often differ from the Western ones, and computed tomography is t available 24/7 in many centres. Other limitations include financial resources and exposure to high doses of radiation. Therefore, some authors promote ultrasound examination as the key tool for the diagsis of acute appendicitis (9), although its actual value remains controversial. According to literature data, its sensitivity varies from 44 to 100%, and specificity from 47 to 99% (19). The discrepancies result, first of all, from the expertise and experience of the ultrasound specialist, but also from the patient-dependent factors such as obesity, large volume of intestinal gases or previous laparotomies. The data confirm our observation that the accuracy of the diagsis highly depended on the radiologist examining the patient. Our study has shown that scoring systems are of limited value in diagsing AA. The majority of the systems showed high sensitivity and positive predictive value allowing to select actually ill patients and to limit unnec-

5 The value of scoring systems for the diagsis of acute appendicitis 69 essary appendectomies. However, occasional very low specificity and negative predictive value are associated with delayed diagsis, and consequently with the increase in the percentage of perforations, diffuse peritonitis and further consequences. In our study, the Tzakis system appears to be most useful and can be considered statistically significantly correlated at the level from 13%. Our results differ from those provided in the literature, particularly as regards the specificity, for which much better results have been achieved. The difference may result, among other things, from the size of study groups. In a large systematic review evaluating the validity of Alvarado score and involving 3,000 patients, the sensitivity was 82% and specificity 81% (with the cut-off point of 7 for the diagsis of AA) (19). The sensitivity and specificity values declared by the systems authors are 91% and 81% for the RIPASA score (10), 95% and 97% for the Tzakis score (9), and 73% and 87% for the Fenyo score (11). Comparisons of two or, less frequently,three scoring systems arecommon in the literature. Comparisons involving more systems, like in our study, are very rare. The Alvarado system is most popular and most frequently compared. It is worth ting that the development of a new scoring system is a purely mathematical procedure, in which point values are assigned depending on distribution of a trait in the examined population. Therefore, it is t surprising that invative scoring systems are always better than those to which they are compared. It seems that the old surgical motto if in doubt take it out becomes irrelevant these days. Additionally, as rightly suggested by Bialas and Gryszkiewicz, due to growing awareness of the patients it may turn out that unjustified appendectomy and exposure to postoperative complications will soon become the subject of indemnification procedures (3). Unfortunately, we are still far from a satisfactory diagstic method and the abundance of the proposed scoring systems adequately illustrates difficulties in finding the perfect one. Some authors suggest that in the age of laparoscopy the quoted motto should be replaced by if they re crook take a look (21). However, the question is whether such laparoscopic evaluation is reliable and whether visually unchanged appendix should be removed (22, 23) references 1. Horzić M, Salamon A, Kopljar M et al.: Analysis of scores in diagsis of acute appendicitis in women. Coll Antropol 2005; 1: Andersen E, Sondenaa K, Soreide JA et al.: Acute appendicitis. Preoperative observation time and diagstic accuracy. Tidsskr Nor Laegeforen 1992; 112: Białas M, Gryszkiewicz M: Podejrzenie ostrego zapalenia wyrostka robaczkowego trudści w rozpoznaniu i metody wspomagające diagstykę. Nowiny Lekarskie 2006; 75: Blomqvist PG, Andersson RE, Granath F et al.: Mortality After Appendectomy in Sweden, Ann Surg ; 233: Alvarado A: A practical score for the early diagsis of acute appendicitis. Ann Emerg Med 1986; 15: Ohmann C, Franke C, Yang Q et al.: Diagstic score for acute appendicitis. Chirurg 1995; 66: Van Way CW 3rd, Murphy JR, Dunn EL et al.: A feasibility study of computer aided diagsis in appendicitis. Surg Gynecol Obstet 1982; 155: Eskelinen M, Ikonen J, Lipponen P: A computerbased diagstic score to aid in diagsis of acute appendicitis: a prospective study of 1333 patients with acute abdominal pain. Theor Surg 1992; 7: Tzanakis NE, Efstathiou SP: A new approach to accurate diagsis of acute appendicitis World J Surg 2005; 29: Chong CF, Thien A, Mackie AJA et al.: Comparison of RIPASA and Alvarado scores for the diagsis of acute appendicitis. Singapore Med J 2011; 52: Fenyo G: Routine use of a scoring system for decision making in a suspected acute appendicitis in adults. Acta Chir Scand 1987; 153: Thuijls G, Derikx JP, Prakken FJ et al.: A pilot study on potential new plasma markers for diagsis of acute appendicitis. Am J Emerg Med 2011; 29: Farooqui W, Pommergaard HC, Burcharth J et al.: The diagstic value of panel of serological markers in acute appendicitis. Scand J Surg 2014 Apr 15. (Epub ahead of print) 14. Andersson M, Rubér M, Ekerfelt C et al.: Can new inflammatory markers improve the diagsis

6 70 D. A. Walczak et al. of acute appendicitis? World J Surg 2014; 38: Schuler JG, Shortsleeve MJ, Goldenson RS et al.: Is there a role for abdominal computed tomographic scans in appendicitis? Arch Surg 1998; 133: Rao PM, Rhea JT, Rattner DW et al.: Introduction of appendiceal CT: impact on negative appendectomy and appendiceal perforation rates. Ann Surg 1999; 229: Jones K, Pena AA, Dunn EL et al.: Are negative appendectomies still acceptable? Am J Surg 2004; 188: McDonald GP, Pendarvis DP, Wilmoth R et al.: Influence of preoperative computed tomography on patients undergoing appendectomy. Am. Surg 2001; 67: Pinto F, Pinto A, Russo A et al.: Accuracy of ultrasography in the diagsis of acute appendicitis in adult patients: review of the literature. Crit Ultrasound J 2013; 5: S Ohle R, O Reilly F, O Brien KK et al.: The Alvarado score for predicting acute appendicitis: a systematic review. BMC Med 2011; 9: Hugh TB, Hugh TH: Appendectomy becoming a rare event. http: //karmak.org/archive/2003/05/ hugh.html 22. Phillips AW, Jones AE, Sargen K: Should the macroscopically rmal appendix be removed during laparoscopy for acute right iliac fossa pain when other explanatory pathology is found? Surg Laparosc Endosc Percutan Tech 2009; 19: Hamminga JT, Hofker HS, Broens PM et al.: Evaluation of the appendix during diagstic laparoscopy, the laparoscopic appendicitis score: a pilot study. Surg Endosc 2013; 27: Received: r. Adress correspondence: Bełchatów, ul. Czapliniecka dr dominikwalczak@gmail.com