Clinical judgment remains of great value in the diagnosis of acute appendicitis

Abstract

Background

Observation and repeated examination may lead to favourable clinical outcomes in the ever-challenging diagnosis of appendicitis. The goal of this study was to evaluate clinical performance in the diagnosis of suspected appendicitis in a centre with limited access to medical imaging technologies and to identify factors associated with complicated cases.

Methods

A retrospective review of the medical records of 211 consecutive surgical cases of suspected appendicitis, spanning an 11-year period, was performed. The delays before treatment and the subsequent patient outcomes were evaluated.

Results

There were 8.1% of cases with negative findings on appendectomy, 75.8% with uncomplicated appendicitis, 12.3% with complicated appendicitis and 3.8% with other surgical conditions. The delay before the first medical consultation was significantly longer in patients with complicated appendicitis. The various delays after the first medical consultation did not differ significantly between the groups.

Conclusions

In the context of limited available medical imaging modalities, clinical observation was not associated with an increased incidence of complicated appendicitis. The presence of complicated appendicitis was associated with the delay before the patient's first medical consultation. Clinical judgment can be prioritized and can lead to good clinical performance in the management of patients with suspected appendicitis, with no significant increase in rates of complicated appendicitis and negative findings on appendectomy.

Appendicitis remains a difficult diagnosis.¹ Physicians have long relied on clinical grounds to make this ever-challenging diagnosis. New diagnostic modalities have not yet been shown conclusively to improve the outcome in terms of negative findings on appendectomy and complicated appendicitis.^2–10 The best results seem to involve algorithms that include some clinical score, even if ultrasonography or computed tomography (CT), or both, are used.^11–13 The best use of ultrasonography and/or CT allows the physician to obtain additional information in equivocal cases.^1,8,14,15

Observation and repeated clinical examination have been evaluated and have led to good clinical outcomes,^12,16–24 whereas a false-negative clinical evaluation carries the potential of higher perforation rates.^1,25 However, perforated and complicated appendicitis are associated more frequently with longer delays before first medical consultation.^2,22,26,27 Negative findings on appendectomy are associated with non-negligible costs and potential morbidity.²⁶

With a lifetime cumulative incidence of 8.6% and 6.7% for men and women, respectively,²⁸ appendicitis is the most frequent abdominal emergency.² Despite long experience with this diagnosis and a crowded literature on the subject, rates of negative findings on appendectomy have not decreased even today.²⁶ Except for a few reports of rates of negative findings on appendectomy below 10%,^22,29–31 most recent studies report rates between 15% and 34%.^{2,7,11,16,26,32–35} The situation is the same regarding the rate of complicated appendicitis, which has not changed substantially over time, remaining between 15% and 30%.^{2,11,22,27,29}

Radiological diagnostic modalities are not consistently available.²² Whereas a sonogram or a CT scan may shorten delay in diagnosis in equivocal cases,^1,31 these imaging modalities can, however, delay appropriate surgery,^7,36 either while a patient is awaiting the examination or because of false-negative results. The importance and value of clinical judgment has been underlined by others.^8,12,23,24 The goal of this study was to evaluate clinical performance in the diagnosis of suspected appendicitis in a centre with limited access to medical imaging technologies and to identify factors associated with complicated appendicitis.

Methods

The operating room registry of the Centre de santé Ste-Famille of Ville-Marie, Quebec, was reviewed to identify all cases with a preoperative and/or postoperative diagnosis of appendicitis. A case was considered

• if the preoperative diagnosis was appendicitis, possibility of appendicitis or abdominal pain of unknown origin;

• if planned surgery was appendectomy or exploratory laparotomy or laparoscopy; and

• if the postoperative diagnosis was appendicitis.

The medical records of all these cases were then individually reviewed. The patients with either a possible or a confirmed diagnosis of appendicitis were then included in the study. The study period was from April 1991 to March 2002. Before this period, all files had been cleared except for administrative information.

Two-hundred and eleven consecutive cases with a preoperative and/or postoperative diagnosis of appendicitis were retrieved. Demographic data, clinical signs, various delays and hospital stay were collected from the records. The initial decision of the surgeon was divided into 2 categories: surgery and observation. The delays were categorized as follows:

(1) delay occurring before first medical consultation (which is related to the patient); and

(2) delay occurring after first medical consultation (which is related to medical care).

The delay after the first medical visit was further divided into

(2.1) delay between first primary care physician visit and surgical consultation;

(2.2) delay between the surgical consultation and the decision to operate; and

(2.3) delay between the decision to operate and the surgical intervention.

The delay before the first medical consultation was estimated from the patient's history of the beginning of symptoms, as noted in the medical records. The first visit to the emergency department was used as the first medical consultation even if the patient had been discharged and had returned for subsequent visits.

The status of the appendix at the time of the operation was obtained from the operative protocol and the pathology report. It was classified as uncomplicated (inflamed, gangrenous without perforation), complicated (perforation, abscess, peritonitis) and normal. Other postoperative diagnoses obtained were divided into surgical and nonsurgical diseases.

Multiple logistic regression analysis was carried out to identify which factors could predict the presence of complicated appendicitis. Multiple stepwise logistic regression, χ² and Student's t test were applied when appropriate. Statistical significance was established at p < 0.05.

Results

The following demographic and clinical data were extracted. There were 82 women (38.9%) and 129 men (61.1%). The proportion of women was slightly higher in the group with negative findings on appendectomy (52.9%), but this difference was not statistically significant compared with their proportions of 37.5% in the group with uncomplicated appendicitis, 38.5% in the group with complicated appendicitis and 37.5% in the group with other pathology. However, the difference attained a statistical significance when the proportion of women who had negative findings on exploration was compared with the rest of the study population (p = 0.047). The mean age of the study population was 25.5 (standard deviation [SD] 14.6) years for the study population. It was 26.5 (SD 10.2) years for the group with negative findings on appendectomy, 24.1 (SD 14.7) years for the group with uncomplicated appendicitis, 29.6 (SD 21.1) years for the group with complicated appendicitis and 32.5 (SD 13.2) years for the group with other pathology. There was no statistically significant difference in age between groups.

Five patients (2.4%) had a comorbidity. Eight patients (3.8%) had other surgical conditions. There were 2 cases of Meckel's diverticulum and 1 case of each of the following conditions: intestinal duplication, perforated carcinoma of the cecum, torsion of the epiploic appendage, pelvic abscess, hemorrhagic oviarian cyst and torsion of an ovarian cyst. Among the remaining 203 patients (96.2%), there were 17 cases (8.4%) of negative findings on appendectomy, 160 cases (78.8%) of uncomplicated appendicitis and 26 cases (12.8%) of complicated appendicitis. For the whole group, the rate of negative findings on exploration was 8.1% (17/211). In these patients, there were 4 cases of adenitis, 2 cases of enteritis, 1 case of typhlitis, 1 case of endometriosis and 9 cases with no diagnosis.

In patients with appendicitis (uncomplicated and complicated), loss of appetite was present in 171 cases (91.9%), nausea and/or vomiting in 139 cases (74.7%), migrating pain in 147 cases (79.0%) and rebound tenderness in 146 cases (78.5%). Mean temperature was 36.8°C (SD 0.6°C) in the group with negative findings on appendectomy, 37.0°C (SD 0.7°C) in the group with uncomplicated appendicitis and 37.2°C (SD 0.8°C) in the group with complicated appendicitis (p = 0.18). Temperature was ≥ 37.5°C in 17.6%, 23.7% and 38.5% of cases in each group, respectively. The mean leukocyte count was 0.010 (SD 0.002) × 10⁹/L in the group with negative findings on appendectomy, 0.015 (SD 0.005) × 10⁹/L in the group with uncomplicated appendicitis and 0.016 (SD 0.004) × 10⁹/L in the group with complicated appendicitis (p < 0.001). It was 0.011 × 10⁹/L or higher in 50.0%, 85.5% and 84.7% of each group, respectively. Sonograms were obtained for 6 patients (2.8%) when a radiologist was present: 1 showed positive findings, 2 showed doubtful findings and 3 showed negative findings.

For the entire group, the mean delay before first medical consultation was 1.6 (SD 1.9) days, and the mean delay between first medical consultation and definitive treatment was 11.0 (SD 9.8) hours. The mean time lapse between the first medical visit and surgical consultation was 5.0 (SD 5.7) hours. The mean delay between surgical consultation and decision to operate was 4.1 (SD 8.2) hours. The mean delay between decision to operate and surgical intervention was 2.0 (SD 2.3) hours. Clinical observation was initially recommended after surgical consultation in 53.0% of cases of negative findings on appendectomy, in 30.0% of uncomplicated cases of appendicitis, in 30.8% of complicated cases, and in 50.0% of patients with other surgical conditions (p = 0.25). For each group, the various delays are presented in Table 1. After multiple stepwise logistic regression, the sole significant factor associated with the presence of complicated appendicitis was the delay before first medical consultation. This finding was given greater weight by the fact that none of the other delays reached statistical significance (Table 1).

Table 1

Diagnostic laparoscopy was carried out in 6 cases (2.8%): 5 cases of uncomplicated appendicitis and 1 case of complicated appendicitis. Table 2 shows the mean postoperative hospital stays, which were significantly longer in patients with complicated appendicitis and in patients with other surgical conditions. The difference in postoperative hospital stay in patients with negative findings on appendectomy compared with patients with uncomplicated appendicitis was not statistically significant. No deaths had occurred in this retrospective study. Three patients were readmitted to hospital for abscess, and all of these were complicated cases.

Table 2

Discussion

Ville-Marie is a small town on the western border of the province of Quebec with a population of 17 000. General surgery and anesthesiology represent the only permanent attending specialties. A radiologist is available 1 or 2 days each month and only during the daytime. The nearest hospital is 150 km away. The practitioners in such a centre must rely on clinical judgment when facing the possibility of a diagnosis of appendicitis. Unfortunately, it is not possible to know the exact number of patients who consulted with a possible diagnosis of appendicitis if they were discharged after a negative evaluation by the surgeon. In addition, it cannot be assumed that no patients were lost to follow-up after a first consultation. This constitutes a potential limitation of this study. However, considering the low incidence of appendicitis and the great distance to the next centre, the number of patients who presented first at this centre and consulted in another centre thereafter is probably low. A retrospective study of the patients with possible appendicitis in this setting may thus represent a good model with which to evaluate the clinical performance of the team.

An 8.1% rate of negative exploration represents a good clinical performance, because a rate of 15% is still considered acceptable.³⁷ Despite many trials to improve these results, it has become apparent that, in most units, the rate of normal appendix removal remains around 15%.^1,2,7,33,34 Unlike previous reports,³⁸ recent reviews have found no link between the frequency of perforation and misdiagnosis.² In the present review, a rate of complicated appendicitis (perforated and/or with peritonitis or abscess) of 12.3% overall also represents a good performance when compared with the rates in the recent literature of between 15% and 30%.^{2,11,22,27,29}

In this review and others,^2,22,26,27 the presence of complicated appendicitis is related to factors controlled by the patients as represented by a significantly longer delay before the first medical consultation. The observation of patients with repeated clinical evaluations was of great value, because no other delays showed statistically significant differences between groups (Table 1). Clinical observation led to longer preoperative delays in cases of normal appendix and other surgical conditions (Table 1). Even though the difference in these delays was not significant, it underlines the suspicion of other processes than appendicitis.

It is possible that ultrasonography and/or CT could have identified appendicitis earlier, as suggested elsewhere.^{1,14,30,31,34} However, false-negative results may increase the rate of complicated cases.^7,25 The problem resides in the fact that patients with persisting pain and negative findings on investigation cannot be discharged without great confidence that a dangerous process is not ongoing. Today, laparoscopy may aid in the management of these patients.^8,22 However, the best recent results incorporate various clinical scores,^11–13 demonstrating that diagnostic imaging modalities should be reserved for equivocal cases or to rule out other diagnosis such as a gynecological condition.^1,8,14 Clinical judgment and repeated physical examinations still have an essential and valuable role in the diagnosis of appendicitis.^8,15,38

In this review, the hospital stay was not longer for patients with a negative exploration, contrary to the findings of other authors.^25,33 The delay was the same for patients with a normal appendix as for patients with uncomplicated appendicitis (Table 2). The significantly longer delays were for patients with complicated appendicitis and patients with other surgical pathologies. The length of hospital stay for patients with negative exploration may be the result not only of the surgical intervention itself but also of the underlying disease.

This study was undertaken to demonstrate that clinical judgment remains valid in the diagnosis of appendicitis by using data from a centre that represents a good natural set-up. The limited availability of medical imaging technologies and confinement to a rural region make clinical evaluation a primary diagnostic tool. This review is from a different, but similar, centre from that reviewed in a previous study²² but reveals the same observations and allows the same conclusions to be drawn.

The results of this study suggest that in the context of limited available medical imaging modalities clinical observation was not associated with an increased incidence of complicated appendicitis. The presence of complicated appendicitis was associated with the delay before the patient's first medical consultation. In conclusion, clinical judgment can be prioritized and can lead to good clinical performance in the management of patients with suspected appendicitis, with no significant increase in rates of complicated appendicitis and negative findings on appendectomy.

Competing interests: None declared.

Correspondence to: Dr. Eric Bergeron, 3120 blvd. Taschereau, Greenfield Park QC J4V 2H1; eric.bergeron@traumaquebec.org