Acute appendicitis is a common cause of acute abdominal pain requiring treatment in patients presenting at the emergency department (ED). EDs screen a large number of patients with abdominal pain, of which a subset is thought to have possible appendicitis and surgical intervention is indicated. Its atypical presentation is similar to many non-surgical disorders which makes it difficult to diagnose. Delayed diagnosis is associated with an increased rate of perforation, more surgical complications, higher medical expenses and sometimes more legal problems.1 2

The management by emergency physicians evaluating patients with suspected appendicitis includes observation, serial laboratory tests and diagnostic imaging. Emergency physicians have to establish a convincing diagnosis before requesting surgical consultation. CT scanning has been reported to have a high rate of diagnostic accuracy in evaluating patients with suspected appendicitis. Helical CT scanning has been reported to have a high sensitivity (range 90–100%), high specificity (range 91–99%) and a high rate of accuracy (range 94–98%).3^–5 Some authors have recommended routine use of CT scanning for patients with acute right lower quadrant pain to improve care and reduce the use of hospital resources in patients suspected of having acute appendicitis.6 However, other studies have shown that preoperative abdominal CT scanning does not improve the accuracy of a diagnosis of appendicitis.7 8 A retrospective study by Flum et al2 revealed that the diagnosis of appendicitis has not improved with the availability of advanced diagnostic testing.

In 2004, most hospitals in Taiwan adopted the global budgeting scheme as part of the national health insurance system. Hospitals in the scheme were no longer able to receive reimbursement from the Bureau of National Health Insurance (BNHI) if they were over budget. CT scanning is still a relatively expensive diagnostic tool and its use is strictly monitored by hospital administration and the BNHI. The use of CT scanning to diagnose appendicitis is controversial as the benefits under a hospital global budgeting scheme are unclear. The purpose of this study was to analyse the costs of CT scanning for the diagnosis of patients admitted to an ED with possible acute appendicitis.

METHODS

This was a retrospective chart review study. Data from patients who were treated at a 1732-bed urban medical centre with annual ED visits of >110 000 were collected from 1 January to 30 June 2005, which included patients admitted to the ED with possible acute appendicitis and those who received emergency appendectomy with or without CT scanning. Patient demographic data, present illness, working diagnosis, laboratory data, ED stay, length of hospital stay and expenses in the ED were reviewed. Operation findings and CT reports (if received) were also reviewed.

The possible appendicitis was categorised in the ED into: (1) acute appendicitis; (2) acute appendicitis should be ruled out; and (3) differential diagnosis including acute appendicitis. It was assumed in this study that the possibility of appendicitis was related to the strength of the clinical presentation of acute appendicitis. CT reports were categorised into: (1) negative; (2) confirmed or suspected appendicitis; and (3) others. “Appendectomy not required” was defined as a change in the preoperative decision after receiving the results of the CT scan which indicated a non-surgical treatment strategy.

We used the reimbursed amount as an approximation to the real cost. The cost analysis included the costs of operation, CT scan, ED stay and hospital stay as well as costs saved by not having to perform an appendectomy or CT scan. We assumed that changes in appendectomy decision were based on the CT findings (“appendectomy not required”). The reimbursement per patient from the BNHI for an abdominal CT scan with and without contrast was New Taiwan dollar (NT)$5035 in 2005 and payment per case for an appendectomy at medical centres was NT$35 125. The average reimbursement from the BNHI per hospital day was about NT$7400 in the same year, and the average total cost of nurses’ and physicians’ salaries per month in this hospital was NT$12 253 265. The average total duration of ED patient stay was 35 142 h/month. We therefore defined the average cost of ED stay per patient per hour as the cost of nurses and physicians divided by the total time of ED patient stay, which was NT$420 in 2005. The reimbursement from the BNHI for appendectomy was NT$7301, while the general anaesthesia fee was NT$3850. The total operation fee was NT$11 151. Using the Diagnosis Related Groups (DRG) system, the cost of appendectomy was NT$35 416 (upper limit NT$42 836). The average exchange rate for NT$ and US$ was 32.14 to 1 in 2005.

RESULTS

A total of 266 cases of abdominal pain with possible acute appendicitis were reviewed. Appendectomy was performed in 207 cases in the ED during the study period. The numbers of patients in the different categories of ED impression, CT reports, operative findings and cases of non-appendectomy are shown in fig 1. An impression of “acute appendicitis” was found in 56/266 cases (21%), “acute appendicitis should be ruled out” was found in 182 cases (68%) and “differential diagnosis including acute appendicitis” in 28 cases (11%). CT scanning was performed in 18%, 60% and 71% of these three groups, respectively. For non-appendectomy cases, an impression of “acute appendicitis” was found in 10% of patients, while 35% had an impression of “acute appendicitis should be ruled out” and 50% had an impression of “differential diagnosis including acute appendicitis”.

• Download figure
• Open in new tab
• Download powerpoint

Figure 1 Numbers of patients in different emergency department impression categories, CT reports, operation findings and cases not requiring appendectomy. Ac.app, acute appendicitis; R/O app, acute appendicitis should be ruled out; DDx app, differential diagnosis including acute appendicitis.

The clinical presentations of the patients who did and did not have a CT scan are shown in table 1. Those who did not have a CT scan had a significantly higher percentage with right lower quadrant pain on presentation than those who did undergo CT scanning. The sex, age, laboratory data, ED stay, hospital stay and costs in the ED in patients who underwent appendectomy, with or without a CT scan, are shown in table 2. Patient age, concentration of high sensitivity C-reactive protein (hsCRP), ED stay, ED expenses and hospital stay were lower in those who did not have a CT scan than in those who did. However, the white cell count was higher in patients who did not have a CT scan than in those who underwent CT scanning. The percentage of non-appendicitis was 7.1% in the group who had a CT scan and 8.4 % in those who did not (p = 0.906).

The cost measurements in the two groups are shown in table 3. The average ED stay and hospital stay were 7.4 h and 5.0 days, respectively, in the group with a CT scan compared with 4.3 h and 3.5 days in the group without a CT scan. The total expenses in the 6-month period of the study were about NT$5.7 and NT$4.9 million in those with and those without a CT scan, respectively. The net costs for patients who received a CT scan were NT$40 728 per patient, which was nearly equal to the net costs for patients who did not receive a CT scan (NT$39 192).

DISCUSSION

In this study we found that patients who received a CT scan were more likely to have uncertain clinical impressions. Patients with atypical presentation had the highest CT scan performance rate compared with the other presentation groups. “Pain over the right lower quadrant” was the most sensitive symptom for diagnosis of acute appendicitis compared with the other symptoms. However, the use of “pain over the right lower quadrant” for preoperative decision-making was limited. For atypical patients it is essential for the emergency physician to order tests with a high sensitivity and accuracy in order to provide better quality of care and to make more efficient use of limited time and resources. CT scanning is a well known and effective diagnostic tool in suspected appendicitis.6 9^–12 Its high accuracy in diagnosing and ruling out acute appendicitis may decrease rates of negative appendectomy as well as unnecessary delays for observation.

In this review, interpretations of CT images were found to be performed faster by an emergency physician or surgeon than by a radiologist. This explains why operations were carried out in patients with negative CT reports. Emergency physicians order CT scans based on their clinical findings or in consultation with a surgeon. Surgeons make the decision to operate based on their own physical examination and laboratory report findings, as well as the results of image studies, if performed. Although the CT scans had a high sensitivity and specificity, there were some false negative diagnoses in patients whose appendicitis was in the early inflammatory stage and whose laboratory and physical examination reports indicated possible appendicitis.

Some reports have suggested that a CT scan is useful for the diagnosis of acute appendicitis in women and children.6 13 17 For women with lower abdominal pain or right lower quadrant pain it is difficult to differentiate acute appendicitis from pelvic inflammatory diseases such as endometritis, salpingitis, salpingo-oophoritis, tubo-ovarian abscess and pelvic peritonitis. The higher negative appendectomy rate in girls and adolescent females is most likely related to the gynaecological sources of pain after puberty that often mimic appendicitis. However, in our review the rate of CT scanning was not related to gender, which is consistent with a similar finding reported by Neumayer et al.18 In our ED, emergency physicians consult with gynaecologists and general surgeons before deciding whether or not to order a CT scan in young or adult women with suspected appendicitis with an atypical presentation. Gynaecologists may perform sonograms to rule out the possibility of pelvic inflammatory disease and, if there is no definite finding, an abdominal CT scan is ordered. There were three women in our study whose operative findings revealed pelvic inflammatory disease in group who did not have a CT scan. The patients who had a CT scan were significantly older, perhaps because clinical symptoms and laboratory data tend to be more atypical in older patients with acute appendicitis due to an age-related decline in pain sensation and immunity response. Ultrasonography is another tool which can be applied in the diagnosis of appendicitis, but there are several disadvantages including dependency on the physician’s experience and less sensitivity in early appendicitis. In addition, obese patients and patients with a retrocaecal appendix are difficult to examine with ultrasonography. It is therefore not used as a standard diagnostic tool for suspected appendicitis in our ED.

Because the use of CT scanning is monitored by the hospital administration and the BNHI, emergency physicians are required to make the most effective use of abdominal CT. In addition, physicians need to consider the risks involved with delayed diagnosis such as possible appendix perforation. Traditional methods of diagnosis in patients with suspected appendicitis can be time-consuming because of the need to conduct laboratory tests. Abdominal CT scans have the potential to shorten the time needed to diagnose the condition and therefore decrease the length of stay in the ED.19 20 A stay for observation in the ED of >12 h results in a cost equal to the cost of a CT scan. Longer stays in the ED for suspected appendicitis tend to have a deleterious effect on patient care, partly because of the potential risk of ruptured appendix occurring during observation. The timing of an abdominal CT scan is therefore important for emergency physicians. If the indication for a CT scan is not compelling, the decision to order a scan might be criticised. Conversely, if a CT scan is indicated, there is a risk of a delayed diagnosis and the additional workload could worsen crowding in the ED.

The length of hospital stay was higher in the group who had a CT scan than in those who did not. This might have been because more older patients received a CT scan due to atypical laboratory results and a presentation which may have been related to underlying diseases such as hypertension or diabetes. One benefit of performing a CT scan is the potential saving of the cost of an unnecessary appendectomy. This is particularly beneficial in patients who would otherwise incur a high cost of postoperative care. Net costs per patient and negative appendicitis rates were not significantly different between the groups with and without a CT scan. In hospitals with a limited budget, the routine use of CT scanning for diagnosis of acute appendicitis is not therefore recommended despite its high rate of accuracy. A number of studies have suggested that the routine use of CT scanning for diagnosis of acute appendicitis is unnecessary.21 22

This study had several limitations. First, the impression of diagnosis was based almost entirely on the physician’s subjective evaluation. We did not calculate the rate of misdiagnosis. The appendectomy included both the conventional method and the laparoscopic method, which resulted in different lengths of hospital stay. We did not adjust for the effects of age and longer length of ED stay, including waiting time for a CT scan in the group with CT. These might decrease some costs in this group.

We conclude that history taking, physical examination and laboratory tests are still useful and cost-effective methods for diagnosing acute appendicitis. CT scanning is recommended for suspected appendicitis in older patients and patients with an underlying disease.