Abstract
Background
In patients with right iliac fossa pain, the need for surgery is largely determined by the likelihood of appendicitis. Patients often undergo ultrasound scanning despite a low diagnostic accuracy for appendicitis. This study aimed to determine the feasibility of a larger trial of computed tomography in the evaluation of patients with atypical right iliac fossa pain.
Materials and methods
A single-centre, unblinded, parallel randomised controlled trial of computed tomography in the assessment of patients with atypical right iliac fossa pain. After a retrospective evaluation, standard care was defined as serial examination with or without ultrasound. Atypical right iliac fossa pain was defined as no firm diagnosis after initial senior review. Simple descriptions of the risks and benefits of computed tomography were devised for patients to consider. Primary objectives were to assess feasibility and acceptability of the study procedures.
Results
A total of 71 patients were invited to participate and 68 were randomised. Final analysis included 31 participants in the standard care arm and 33 in the computed tomography arm, with comparable demographics. Computed tomography was associated with superior diagnostic accuracy, with 100% positive and negative predictive value. The proportion of scans that positively influenced management was 73% for computed tomography and 0% for ultrasound. In the computed tomography arm, there was a trend towards a shorter length of stay (2.3 vs 3.1 days) and a lower negative laparoscopy rate (2 of 11 vs 4 of 9).
Conclusion
A large randomised trial to evaluate the use of unenhanced computed tomography in atypical right iliac fossa pain appears feasible and justified.
Keywords: Computed tomography, x-ray, Abdominal pain, Ultrasound, Appendicitis, Appendicectomy
Introduction
Appendicectomy is the most commonly performed operation in emergency general surgery.1,2 Migratory right iliac fossa pain is the typical presentation of appendicitis, yet the majority of patients with such pain can be managed conservatively.3,4 In this patient group, the need for surgery is largely determined by the likelihood of acute appendicitis.
Recent data illustrate the wide variation in rates of laparoscopy.5 A lower threshold for laparoscopy appears to be associated with higher negative appendicectomy rates, while lower rates of laparoscopy may risk a higher rate of perforated appendicitis. Overall, the rate of negative appendicectomy remains over 20%. In young adults, laparoscopic appendicectomy is now a safe and widely accepted procedure, but is not without risk.6,7
The role of diagnostic imaging in this setting is also contentious. In the UK, ultrasound scanning remains widely used as the first line of imaging in young adults with equivocal right iliac fossa pain, yet the appendix is rarely identified.8 The diagnostic accuracy of computed tomography (CT) in acute appendicitis is established, but concerns around radiation exposure and cost have limited its use.9–12 It may now represent an effective means with which to identify which patients need operative management.
The IMPACT (Imaging Possible Appendicitis with CT) trial was designed to evaluate the feasibility of a study to assess CT in patients with atypical RIF pain.
Materials and methods
A single-centre, parallel design feasibility randomised controlled trial of CT as first-line imaging in atypical right iliac fossa pain was performed at James Cook University Hospital. The trial management group consisted of surgical trainees, a surgical consultant, radiologists and a medical physicist. The study was sponsored by South Tees Hospitals NHS Foundation trust, approved by the Preston Research Ethics Committee (16/NW/0491) and registered with ClinicalTrials.gov (NCT03570398). The trial was designed in light of previous data from our unit. In a retrospective analysis, 54 adults were admitted with atypical right iliac fossa pain and ultrasound was performed in 36. Based on these data, standard care was defined as serial clinical assessment with or without ultrasound.
Eligibility
Adult patients (18–60 years) presenting with acute right iliac fossa abdominal pain and/or tenderness were considered for inclusion. Exclusion criteria included pregnancy and previous appendicectomy. Patients booked for appendicectomy after the first registrar or consultant review were also excluded, to rule out those with a convincing clinical diagnosis of appendicitis. Atypical right iliac fossa pain was defined as cases in which right iliac fossa pain was a prominent presenting symptom and after review by a surgical registrar or consultant, there was no firm diagnosis and no plan for operative management.
Randomisation
Eligible patients were identified and invited to participate by on-call surgical trainees and consultants. Patients were informed of the study, provided a participant information sheet and allowed a minimum of 30 minutes to consider participation. After informed consent was obtained, online one to one block randomisation was performed using a secure password-protected system. Computer-generated blocks of size 6 were used to randomise participants to standard care or CT. Patients who declined to participate were also recorded. In the context of an emergency presentation to hospital, patients who declined to participate were not asked to justify their decision.
Imaging intervention
Patients in the standard care group had an ultrasound requested at the time of randomisation. The protocol did not mandate that the scan be performed on an in-patient basis. Patients in the CT arm had an urgent in-patient unenhanced CT requested. CT was performed on an inpatient basis and, in those participants who underwent surgery, as a preoperative test. Standardised reporting templates for ultrasound and CT were embedded within the hospital’s radiology reporting system (Appendix 1). After randomisation and scan request, all further clinical management was at the discretion of the responsible clinical team.
CT was performed as a single breath-hold helical scan from the superior endplate of the T12 to the symphysis pubis. All scans were performed on a Somatom Definition AS scanner (Siemens AG, Munich, Germany). Axial slices were spaced at 3-mm intervals, with 3-mm coronal and sagittal reconstructions. Based on national reference data and dose reports from our own archived scans, the maximum total protocol dose was calculated at 5 mSv. This estimate was used to inform the wording of the patient information sheet. The risk of developing a cancer as a result of the study CT scan was quoted as 3 in 10,000. The effect on fertility in women of childbearing age was deemed to be negligible. All scans were reported by the duty radiology registrar and reviewed by the duty consultant.
Ultrasound scans were performed by radiologists and sonographers in accordance with the trust protocol for a transabdominal abdominopelvic ultrasound. For both CT and ultrasound, visualisation of the appendix was recorded. Radiologists and sonographers reported the finding of appendicitis as present, absent or indeterminate based on predetermined criteria (Appendix 1).
Endpoints and definitions
The primary outcome was acceptability of unenhanced CT in patients with atypical right iliac fossa pain. This was assessed by the acceptability of the study protocol to patients, as measured by rate of refusal to participate. Secondary outcomes were time to scan report, negative laparoscopy rate and length of stay. Imaging tests were assessed in terms of diagnostic accuracy (positive and negative predictive values) and their contribution to clinical management. The acceptability of the protocol to consultant surgeons was evaluated based on study recruitment.
A diagnosis of appendicitis required histopathological confirmation. The absence of appendicitis was defined as negative histology or clinical resolution without antibiotics. For each ultrasound and CT scan, the diagnostic accuracy for appendicitis was assessed. An accurate radiological diagnosis required positive identification of the appendix and an accurate assessment of the presence or absence of inflammation. A scan was deemed to have contributed to patient care when the findings could be used to positively inform management. Negative laparoscopy was recorded when there was no clear benefit from surgical intervention.
Data collection
Study recruitment was actively monitored as the study progressed. After recruitment was complete, study data were recorded from paper notes, electronic records, radiology reports and pathology reports. The clinical probability of appendicitis was evaluated using the MANTRELS score, so as to characterise the clinical appearance of the trial participants.13 Readmissions within two weeks were recorded. Cost analysis was performed in collaboration with the trust finance team based on current NHS tariffs.
Statistical analysis and reporting
Owing to the nature of this feasibility study, no formal sample size calculation was performed. Recommendations for good practice suggest that 20–30 patients per treatment group should provide sufficient data to assess the feasibility of a trial, investigate the distribution of outcome measures and estimate with adequate precision standard deviations of key study parameters.14
Minitab 17 was used for statistical analysis. Categorical data were analysed using chi square cross-tabulation, and Fisher’s exact test was used for individual cell counts less than five. Continuous data were analysed using the Mann-Whitney U-test. The study was reported in compliance with the CONSORT extension for pilot and feasibility studies.
Results
A total of 71 eligible patients were invited to participate in the IMPACT study, of whom 68 agreed to participate (fig 1). All 11 consultant general surgeons in the unit contributed to recruitment. The demographics of the two groups were balanced (Table 1). Only four participants were aged over 50 years.
Figure 1.
CONSORT diagram.
Table 1.
Data summary for standard care and computed tomography groups.
| Computed tomography (n = 33) | Standard care (n = 31) | |
| Demographics: | ||
| Female : male | 26 : 7 | 23 : 8 |
| Age, median (IQR) | 28 (23–36) | 29 (27–40) |
| Body mass index, median (IQR) | 26.9 (24.4–36.2) | 29.0 (25.3–34.0) |
| Outcomes: | ||
| Inpatient scans, n (%) | 33 (100%) | 23 (74%) |
| Time from admission to scan, median days (IQR) | 0.15 (0.07–0.45) | 1.1 (0.9–1.8) |
| Time from admission to reported scan, median days (IQR) | 0.6 (0.2–0.9) | 1.1 (0.9–1.8) |
| Time from admission to decision to operate, median days (IQR) | 0.9 (0.4–1.5) | 1.2 (1.0–2.0) |
| Inpatient stay, median days (IQR) | 2.3 (1.2–3.0) | 3.1 (1.9–4.0) |
| Laparoscopy performed, n (%) | 11 (33%) | 9 (29%) |
| Appendicitis confirmed, n (%) | 8 (24%) | 4 (13%) |
| Negative laparoscopy (%) | 2/11 (18%) | 4/9 (44%) |
| Appendix positively identified (%) | 26/33 (79%) | 1/23 (4%) |
| Scans contributing to decision making (%) | 24/33 (73%) | 0/23 (0%) |
| Diagnostic accuracy (%) | 24/33 (73%) | 0 (0%) |
| Negative predictive value (%) | 100 | NR |
| Positive predictive value (%) | 100 | 0 |
IQR, interquartile range; NR: not reportable.
In the standard care group, 23 patients had inpatient ultrasound. Four were managed non-operatively and left hospital without imaging. Seven patients in the standard care arm had CT during the same admission. Nine patients in the standard care group underwent laparoscopy. There were four cases of appendicitis and one case of retrograde menstruation.
In the CT group, all 33 patients had inpatient CT. The median acquisition time for the protocol scan was 9.3 seconds and the median dose-length product was 121 mGy-cm. The scans were unenhanced so there were no adverse events related to contrast. Eleven patients underwent laparoscopy. Eight had appendicitis and one had oophoritis with purulent fluid in the pelvis.
Clinical risk stratification was completed for 63 of 64 patients (fig 2). For those patients in whom a diagnosis of acute appendicitis was predicted to be either ‘unlikely’ or ‘very probable’, clinical scoring was accurate in 32 of 34 cases. The majority of patients in the current study were stratified as intermediate risk (‘possible’ or ‘probable’ appendicitis) and clinical risk scoring was less helpful in these groups. The distribution of scores in the two arms was very similar. The median score in the standard care arm was 5, suggesting possible appendicitis. In the CT arm, the median score of 4 indicated appendicitis was unlikely. Appendicitis was ultimately diagnosed in 5 of 52 patients with low-risk scores (unlikely or possible appendicitis).
Figure 2.
The clinical likelihood of appendicitis in study participants, calculated using the MANTRELS score.
Median time from admission to scan report was 0.6 days (interquartile range, IQR, 0.2–0.9) in the CT group. In the standard care group, the time to ultrasound report was 1.1 days (IQR 0.9–1.8). The appendix was positively identified in 26 of 33 scans in the CT group, and in 1 of 23 ultrasound scans in the standard care group. With respect to a diagnosis of acute appendicitis, the accuracy of these scans was 73% (24 of 33) in the CT group and 0% (0 of 23) in the standard care group. When the appendix was positively identified at CT, both the negative and positive predictive value was 100%.
In addition to acute appendicitis, CT findings included diverticulitis and a misplaced contraceptive implant. Two patients were shown to have ileocaecitis, one of whom was subsequently diagnosed with Crohn’s disease.
Median inpatient stay was 2.3 days (IQR 1.2-3.0) in the CT group and 3.1 days (IQR 1.9-4.0) in the standard care group. Negative laparoscopy rate was 2 of 11 in the CT group and 4 of 9 in the standard care group. Across the two groups, there were three cases in which the appendix was macroscopically normal at laparoscopy and no other positive findings were reported. In all cases, the appendix was removed and found to be histologically normal.
There were three readmissions within two weeks. Two occurred in the standard care group and one in the CT group. There were no missed cases of acute appendicitis.
The cost of an ultrasound scan without contrast is £79.59. The CT used in the current study cost £98.40. The cost of an overnight stay on a surgical ward was £340. The average cost of an attendance for abdominal pain including laparoscopy was £3034 (age 19+, without major complications/comorbidity). A non-elective admission with abdominal pain without intervention was costed at £578, so the average total cost of a laparoscopy was calculated as £2,456.
Discussion
Right iliac fossa pain is one of the most common referrals to general surgeons and the diagnosis is often unclear after initial clinical assessment. Acute appendicitis is a key differential as surgery remains the treatment of choice. The diagnostic uncertainty associated with the condition has led to an abundance of studies comparing imaging modalities and the primary endpoint for almost all relates to the diagnostic accuracy of the test in question.
This feasibility study was designed to consider the impact of low-dose CT on the clinical journey of patients with atypical right iliac fossa pain. Patient attitudes towards ionising radiation, the acceptability of CT to surgeons and radiologists and the interpretation of equivocal results are all pertinent to the utility of CT in this setting.
In the current study we developed a plain-English description of the risks of CT and ensured potential participants had time to digest this information. Among eligible patients, the recruitment rate of 96% indicates that patients found the radiation exposure acceptable. Each of the consultant surgeons on the emergency rota recruited patients to the study, and the radiology department supported the work on a cost-neutral basis. Our experience indicates that patients, surgeons and radiologists find the use of CT to be acceptable in this setting.
The safety of the protocol was also evaluated. The unenhanced nature of the scan avoided the risk of contrast-related complications. Follow-up data indicated that there were no missed cases of acute appendicitis. In contrast with ultrasound, CT positively identified the appendix in nearly 80% of cases and in these patients, the negative predictive value was 100%. Despite the lack of contrast and the low radiation dose, a CT that ruled out appendicitis was reliable. This finding was consistent with a previous study in which patients with atypical right iliac fossa pain were randomised to either ultrasound or CT.12
In this small sample, patients randomised to a CT scan tended to earlier operations, fewer negative laparoscopies and shorter hospital admissions. CT identified appendicitis in patients assessed clinically to be at low risk and excluded appendicitis in patients assessed to be at high risk. As a pilot study there was insufficient power to address these issues comprehensively, but the use of CT may have a positive impact on clinical decision making.
Patients with a clinical presentation of probable appendicitis were excluded from this study. The authors support the use of laparoscopy in this group, without further imaging. Conversely, there is a group of patients who present with right iliac fossa pain who do not require admission. The single acquisition scan time for the CT protocol used in the current study was 9 seconds. Without a prescription for contrast, the scan did not require active input from a radiologist. Such an investigation may be incorporated into the initial assessment of a patient on a surgical assessment unit.
Only four patients were aged over 50 years. This may reflect clinicians’ reluctance to bypass CT in this older group. In a larger trial, limiting the age range to a younger cohort would help to focus the study design.
In comparison with other countries, the use of CT in young adults with right iliac fossa pain appears limited in the UK.15 Concerns around ionising radiation are commonly cited but are not reflected in the participation rate in the current study. Analysis of the radiation doses to which study participants were exposed suggested an overestimate in our dose predictions. The median dose-length product of 121 mGy-cm in this study is consistent with a risk of 1:10,000 of developing a new cancer, three times lower than the risk suggested in the participant information sheet.
The potential benefits of CT include a lower negative laparoscopy rate, minimising the risk of operative complications. In the current study, surgeons removed a macroscopically normal appendix in the absence of other positive findings, a practice reflected in the literature.4 Complication rates for laparoscopy in non-specific right iliac fossa pain would therefore be expected to approximate those for laparoscopic appendicectomy. In two large randomised controlled trials comparing laparoscopic and open appendicectomy, the complication rates were 8% (24/282) and 19% (47/244) in the laparoscopic arms. A single postoperative death was recorded in the larger series.6,7 Negative laparoscopy rate would represent a potential efficacy outcome for a larger study into the role of CT in atypical right iliac fossa pain.
As in previous reports, the appendix was rarely identified on ultrasound.8 In the one case of appendicitis identified on ultrasound, histopathology was reported to be normal. There were no cases in which a laparoscopy was avoidable on the basis of the ultrasound result. Conversely, CT identified significant pathology in addition to appendicitis.
In the design phase of the current study, we considered the use of magnetic resonance imaging as an alternative diagnostic modality in patients with right iliac fossa pain. The benefits of magnetic resonance imaging (MRI) include high diagnostic accuracy and the avoidance of ionising radiation.16 The limitations relate largely to availability and cost. Gastrointestinal MRI is not a core competency for radiologists or radiographers in the UK. As a result, abdominal MRI is not consistently available out of hours, even in larger centres. This limits its role as an emergency investigation.
The limited sample size in this study precludes any detailed cost projections. In our experience, patients in the CT arm appear to leave hospital nearly a day earlier than those managed by standard care. On average this would equate to a saving of over £270 per patient, which far exceeds the added cost of CT over ultrasound (£19). Our data indicate that the use of CT in patients with atypical right iliac fossa pain would also reduce the negative laparoscopy rate, although the numbers were small. In our analysis, the cost of a negative appendicectomy was around £2456, comparable to the figure of £2355 quoted in a recent study on the use of laparoscopy in suspected appendicitis.17 Based on the savings relating to length of stay and laparoscopy rate, the current data indicate that a policy of routine CT in patients with atypical right iliac fossa pain would be cost-effective.
The major limitations of the current study relate to the small sample size and the single-centre design. In our unit, patients wait longer for abdominal ultrasound than for a CT. This skews the results around the timing of decision making and length of stay. Nonetheless, ultrasound is a labour-intensive test and we do not believe these delays are unique to our unit. Additionally, increasing the resources available for ultrasound would not improve the rate with which the appendix was positively identified.
This small study of patients with atypical right iliac fossa pain suggests that the use of CT is safe, acceptable and potentially beneficial to patient care. Further multicentre evaluation of CT in this setting is warranted.
Acknowledgements
The authors are grateful to Sarah-Louise Chapple for her support as a medical physics expert. The following contributors recruited patients to the study and gave written agreement for their inclusion: Ish Ahmed, Rehaan Ansari, Lucy Bookless, Nnaemeka Chidumije, Jacob Duffin, Nehemiah Edwards, Mihai Firescu, Stephanie Grainge, Andrew Harbit, Andrew Hollingsworth, Zehra Imam, Laura Keast, Mimi Li, Simon Mbarushimana, Phil McElnay, Jamie Mellen, Dan Miller, Chakri Munipalle, Aya Musbahi, David Norton, Kareem Omar, Ben Smith, Alvin Teo, Jianan Yuan.
References
- 1.Flum DR, Koepsell T. The clinical and economic correlates of misdiagnosed appendicitis: nationwide analysis. 2002; (7): 799–804. [DOI] [PubMed] [Google Scholar]
- 2.Hall MJ, DeFrances CJ, Williams SN et al. National Hospital Discharge Survey: 2007 summary. Natl Health Stat Report 2010; : 1–20, 24. [PubMed] [Google Scholar]
- 3.Simmen HP, Decurtins M, Rotzer A et al. Emergency room patients with abdominal pain unrelated to trauma: prospective analysis in a surgical university hospital. Hepatogastroenterology 1991; (4): 279–282. [PubMed] [Google Scholar]
- 4.Humes DJ, Simpson J. Acute appendicitis. BMJ 2006; (7567,): 530–534. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.National Surgical Research Collaborative. Multicentre observational study of performance variation in provision and outcome of emergency appendicectomy. Br J Surg 2013; (9): 1,240–1,252. [DOI] [PubMed] [Google Scholar]
- 6.Hellberg A, Rudberg C, Kullman E et al. Prospective randomized multicentre study of laparoscopic versus open appendicectomy. Br J Surg 1999; (1): 48–53. [DOI] [PubMed] [Google Scholar]
- 7.Pedersen AG, Petersen OB, Wara P et al. Randomized clinical trial of laparoscopic versus open appendicectomy. Br J Surg 2001; (2): 200–205. [DOI] [PubMed] [Google Scholar]
- 8.Stewart JK, Olcott EW, Jeffrey RB. Sonography for appendicitis: nonvisualization of the appendix is an indication for active clinical observation rather than direct referral for computed tomography. J Clin Ultrasound 2012; (8): 455–461. [DOI] [PubMed] [Google Scholar]
- 9.Anderson BA, Salem L, Flum DR. A systematic review of whether oral contrast is necessary for the computed tomography diagnosis of appendicitis in adults. Am J Surg 2005; (3): 474–478. [DOI] [PubMed] [Google Scholar]
- 10.in't Hof KH, van Lankeren W, Krestin GP et al. Surgical validation of unenhanced helical computed tomography in acute appendicitis. Br J Surg 2004; (12): 1,641–165. [DOI] [PubMed] [Google Scholar]
- 11.Rud B, Oafsson L, Vejborg T et al. Diagnostic accuracy of computed tomography for appendicitis in adults. Cochrane Database Syst Rev 2012; (7): CD009977. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Horton MD, Counter SF, Florence MG, Hart MJ. A prospective trial of computed tomography and ultrasonography for diagnosing appendicitis in the atypical patient. Am J Surg 2000; (5): 379–381. [DOI] [PubMed] [Google Scholar]
- 13.Alvarado A. A practical score for the early diagnosis of acute appendicitis. Ann Emerg Med 1986; (5): 557–564. [DOI] [PubMed] [Google Scholar]
- 14.Lancaster GA, Dodd S, Fau-Williamson PR, Williamson PR. Design and analysis of pilot studies: recommendations for good practice. J Eval Clin Pract 2004; (2): 307–312. [DOI] [PubMed] [Google Scholar]
- 15.Raman SS, Osuagwu FC, Kadell B et al. Effect of CT on false positive diagnosis of appendicitis and perforation. N Engl J Med 2008; (9): 972–973. [DOI] [PubMed] [Google Scholar]
- 16.Cobben L, Groot I, Kingma L et al. A simple MRI protocol in patients with clinically suspected appendicitis: results in 138 patients and effect on outcome of appendectomy. Eur Radiol 2009; (5): 1,175–1,183. [DOI] [PubMed] [Google Scholar]
- 17.D’Souza N, Marsden M, Bottomley S et al. Cost-effectiveness of routine imaging of suspected appendicitis. Ann R Coll Surg Engl 2018; (1): 47–51. [DOI] [PMC free article] [PubMed] [Google Scholar]