Impact of helical computed tomography on the rate of negative appendicitis

Efficacy of automated machine learning models and feature engineering for diagnosis of equivocal appendicitis using clinical and computed tomography findings

This study evaluates the diagnostic efficacy of automated machine learning (AutoGluon) with automated feature engineering and selection (autofeat), focusing on clinical manifestations, and a model integrating both clinical manifestations and CT findings in adult patients with ambiguous computed tomography (CT) results for acute appendicitis (AA). This evaluation was compared with conventional single machine learning models such as logistic regression(LR) and established scoring systems such as the Adult Appendicitis Score(AAS) to address the gap in diagnostic approaches for uncertain AA cases. In this retrospective analysis of 303 adult patients with indeterminate CT findings, the cohort was divided into appendicitis (n = 115) and non-appendicitis (n = 188) groups. AutoGluon and autofeat were used for AA prediction. The AutoGluon-clinical model relied solely on clinical data, whereas the AutoGluon-clinical-CT model included both clinical and CT data. The area under the receiver operating characteristic curve (AUROC) and other metrics for the test dataset, namely accuracy, sensitivity, specificity, PPV, NPV, and F1 score, were used to compare AutoGluon models with single machine learning models and the AAS. The single ML models in this study were LR, LASSO regression, ridge regression, support vector machine, decision tree, random forest, and extreme gradient boosting. Feature importance values were extracted using the "feature_importance" attribute from AutoGluon. The AutoGluon-clinical model demonstrated an AUROC of 0.785 (95% CI 0.691-0.890), and the ridge regression model with only clinical data revealed an AUROC of 0.755 (95% CI 0.649-0.861). The AutoGluon-clinical-CT model (AUROC 0.886 with 95% CI 0.820-0.951) performed better than the ridge model using clinical and CT data (AUROC 0.852 with 95% CI 0.774-0.930, p = 0.029). A new feature, exp(-(duration from pain to CT)3 + rebound tenderness), was identified (importance = 0.049, p = 0.001). AutoML (AutoGluon) and autoFE (autofeat) enhanced the diagnosis of uncertain AA cases, particularly when combining CT and clinical findings. This study suggests the potential of integrating AutoML and autoFE in clinical settings to improve diagnostic strategies and patient outcomes and make more efficient use of healthcare resources. Moreover, this research supports further exploration of machine learning in diagnostic processes.

Evaluation of the diagnostic performance of a decision tree model in suspected acute appendicitis with equivocal preoperative computed tomography findings compared with Alvarado, Eskelinen, and adult appendicitis scores: A STARD compliant article

This study evaluated the diagnostic performance of a new clinical approach based on decision tree (DT) analysis in adult patients with equivocal computed tomography (CT) findings of acute appendicitis (AA) compared with previous scoring systems.This retrospective study of 244 adult patients with equivocal CT findings included appendicitis (AG, n = 80) and non-appendicitis (NAG, n = 164) groups. The chi-squared automatic interaction detection algorithm was for AA prediction. A receiver operating characteristic curve analysis and area under the curve (AUC) were used to compare the DT analysis with Alvarado, Eskelinen score, and adult appendicitis scores (AAS).The following factors were selected for AA prediction: rebound tenderness severity, migration, urinalysis, symptom duration, leukocytosis, neutrophil count, and C-reactive protein levels. The DT comprised 11 final nodes with the following AA probabilities: node 1, 100% (16/16); node 2, 90% (9/10); node 3, 80% (8/10); node 4, 60.9% (14/23); node 5, 50% (3/6); node 6, 43.8% (7/16); node 7, 22.6% (12/53); node 8, 13% (10/77); node 9, 5.6% (1/18); node 10, 0% (0/12); and node 11, 0% (0/3). The AUC of the DT was higher (0.850 [95% confidence interval {CI}; 0.799-0.893]) than the Alvarado score (0.695 [95% CI; 0.633-0.752]), AAS (0.749 [95% CI; 0.690-0.802]), and the Eskelinen score (0.715 [95% CI; 0.654-0.770]). The results were statistically significant when compared with the AUCs of the Alvarado score, Eskelinen score, and AAS (P < .001, P < .001, P = .003, respectively).The DT-based approach facilitated AA diagnosis and determination of clinical status in patients with equivocal preoperative CT findings and ambiguous results.

Suspected appendicitis pathway continues to lower CT rates in children two years after implementation

BACKGROUND

We implemented a protocol to evaluate pediatric patients with suspected appendicitis using ultrasound as the initial imaging modality. CT utilization rates and diagnostic accuracy were evaluated two years after pathway implementation.

METHODS

This was a retrospective observational study of patients <18 years evaluated for suspected appendicitis. CT rates were compared before and after implementation of the protocol, and monthly CT rates were calculated to assess trends in CT utilization.

RESULTS

CT use decreased significantly following pathway implementation from 94.2% (130/138) to 27.5% (78/284; p < 0.001). Linear regression of monthly CT utilization demonstrated that CT rates continued to trend down two years after pathway implementation. Adherence to the pathway was 89.8% (255/284). Negative appendectomy rate was 2.4% (2/85) in the post-pathway period.

CONCLUSIONS

Adherence to a pathway designed to evaluate pediatric patients with suspected appendicitis using ultrasound as the primary imaging modality has led to a sustained decrease in CT use without compromising diagnostic accuracy.

Can clinical scoring systems improve the diagnostic accuracy in patients with suspected adult appendicitis and equivocal preoperative computed tomography findings?

Objective

Adult appendicitis (AA) with equivocal computed tomography (CT) findings remains a diagnostic challenge for physicians. Herein we evaluated the diagnostic performance of several clinical scoring systems in adult patients with suspected appendicitis and equivocal CT findings.

Methods

We retrospectively evaluated 189 adult patients with equivocal CT findings. Alvarado, Eskelinen, appendicitis inflammatory response, Raja Isteri Pengiran Anak Saleha Appendicitis (RIPASA), and adult appendicitis score (AAS) scores were evaluated, receiver operating characteristic analysis was conducted, and the optimal, low, and high cut-off values were determined for patient classification into three groups: low, intermediate, or high.

Results

In total, 61 patients were included in the appendicitis group and 128 in the non-appendicitis group. There were no significant differences between the area under the curve of the clinical scoring systems in the final diagnosis of AA for equivocal appendicitis on CT (Alvarado, 0.698; Eskelinen, 0.710; appendicitis inflammatory response, 0.668; RIPASA, 0.653; AAS, 0.726). A RIPASA score greater than 7.5 had a high positive predictive value (90.9) and an AAS score less than or equal to 5 had a high negative predictive value (91.7) in the diagnosis of AA.

Conclusion

The accuracy of clinical scoring systems in the diagnosis of AA with equivocal CT findings was moderate. Therefore, a high RIPASA score may assist in the diagnosis of AA in patients with equivocal CT findings, and a low AAS score may be used as a criterion for patient discharge. Most patients presented with intermediate scores. The patients with equivocal CT findings may be considered as a third diagnostic category of AA.

Improvement in the diagnosis of appendicitis

Our attempts to systematically improve accuracy in the evaluation of patients with suspected appendicitis are, in some ways, hindered by the fact that the condition is so frequently straightforward to diagnose. Careful history-taking and physical examination are reliable in most patients. However, establishing the diagnosis with these skills alone remains vulnerable to conditions that masquerade as acute appendicitis. A substantial body of clinical research over the last quarter-century has shown that improved accuracy is possible. Strategies for improvement include the use of diagnostic scoring systems, laboratory makers such as CRP, diagnostic laparoscopy, and advanced imaging modalities such as CT, MRI, and US. How clinicians use these strategies depends on many factors related to practice setting, the population served, and clinical goals. In children, for instance, the desire to limit exposure to ionizing radiation competes with the greater anatomic detail that a CT scan can provide; at the same time, many hospitals that treat children do not have the resources to maintain the sort of full-time, highly sophisticated abdominal US programs that achieve the highest rates of diagnostic accuracy in clinical studies. Trade-offs have to be made, but improvement is possible in almost all groups of patients: the clinical community should no longer settle for a 15% NA rate when 5% is clearly possible without adverse consequences. Many clinicians will be faced with the task of evaluating patients suspected of having acute appendicitis. A deliberate, proactive, and, ideally, benchmarked strategy for improving diagnosis should be the standard to which we hold ourselves and the promise we deliver to our patients.

The negative appendectomy rate: who benefits from preoperative CT?

OBJECTIVE

The purpose of this article is to determine the negative appendectomy rates of patients who did and did not undergo preoperative CT and to determine, more specifically, whether men benefit from preoperative CT.

MATERIALS AND METHODS

We identified 512 patients who had a nonincidental appendectomy between July 1, 2002, and June 30, 2007. Pathology records were compared with a radiology records search to determine which patients underwent preoperative CT. Proportions of patients were compared between groups using the Fisher exact test.

RESULTS

Of 512 patients who had a nonincidental appendectomy, 465 (91%) underwent preoperative CT, and 47 (9%) underwent appendectomy only on the basis of clinical findings. Overall, 22 of 465 patients (4.7%) who underwent preoperative CT had a negative appendectomy compared with six of 47 patients who did not undergo preoperative imaging (negative appendectomy rate, 12.7%; p = 0.03). Among men, six of 237 (2.5%) with preoperative CT had a negative appendectomy, versus five of 42 without imaging (negative appendectomy rate, 11.9%; p = 0.01).

CONCLUSION

The negative appendectomy rate was decreased for adult patients who underwent preoperative CT compared with patients who did not undergo preoperative imaging. Although most prior studies have suggested that CT is efficacious only in decreasing the negative appendectomy rate among women, we found that men benefit from CT as well.

Impact of computed tomography of the abdomen on clinical outcomes in patients with acute right lower quadrant pain: a meta-analysis

BACKGROUND

Clinical evaluation alone is still considered adequate by many clinicians who treat patients with appendicitis. The impact of computed tomography (CT) on clinical outcomes remains unclear, and there is no consensus regarding the appropriate use of CT in these patients. We sought to evaluate the impact of abdominal CT on the clinical outcomes of patients presenting with suspected appendicitis.

METHODS

We conducted a systematic review of the literature to identify studies that examined clinical outcomes related to the use of abdominal CT in the diagnosis of acute appendicitis. Inclusion criteria were studies of adult patients with suspected appendicitis that evaluated the impact of abdominal CT on negative appendectomy rates, perforation rates or time to surgery. Two independent investigators reviewed all titles and abstracts and extracted data from 28 full-text articles. Statistical analysis was conducted using Review Manager 5.0.10 software.

RESULTS

The negative appendectomy rate was 8.7% when using CT compared with 16.7% when using clinical evaluation alone (p < 0.001). There was also a significantly lower negative appendectomy rate during the CT era compared with the pre-CT era (10.0% v. 21.5%, p < 0.001). Time to surgery was evaluated in 10 of the 28 studies, 5 of which demonstrated a significant increase in the time to surgery with the use of CT. Appendiceal perforation rates were unchanged by the use of CT (23.4% in the CT group v. 16.7% in the clinical evaluation group, p = 0.15). Similarly, the perforation rate during the CT era was not significantly different than that during the pre-CT era (20.0% v. 19.6%, p = 0.74).

CONCLUSION

This meta-analysis supports the hypothesis that the use of preoperative abdominal CT is associated with lower negative appendectomy rates. The use of CT in the absence of an expedited imaging protocol may delay surgery, but this delay is not associated with increased appendiceal perforation rates. Routine CT in all patients presenting with suspected appendicitis could reduce the rate of unnecessary surgery without increasing morbidity.

Update on imaging for acute appendicitis

Acute appendicitis is a common surgical emergency and the diagnosis can often be made clinically; however, many patients present with atypical findings. For these patients, there are multiple imaging modalities available to aid in the diagnosis of suspected appendicitis in an effort to avoid a negative appendectomy. Computed tomography is the test of choice in most patients in whom the diagnosis is not certain. Ultrasonography is particularly useful in children and pregnant women. Magnetic resonance imaging is recommended when ultrasonography is inconclusive. Appropriate use of these imaging studies avoids delays in treatment, prolonged hospitalization, and unnecessary surgery.

Making the diagnosis of acute appendicitis: do more preoperative CT scans mean fewer negative appendectomies? A 10-year study

PURPOSE

To determine the frequency of preoperative computed tomography (CT) in the evaluation of patients suspected of having appendicitis at one institution during the past 10 years and to determine whether changes in CT utilization were associated with changes in the negative appendectomy rate.

MATERIALS AND METHODS

Institutional review board approval was obtained, and a waiver of informed consent was granted for this HIPAA-compliant study. A surgical database search yielded medical record numbers of 925 patients (526 [ 56.9%] men and 399 [43.1%] women; mean age, 38 years (range, 18-95 years]) who underwent urgent appendectomy between January 1998 and September 2007. Patients who were younger than 18 years of age at the time of surgery were excluded. CT, pathology, and surgery reports were reviewed. By using logistic regression, changes in the proportion of patients undergoing CT and in the proportion of patients undergoing each year appendectomy in which the appendix was healthy were evaluated. Subgroup analyses based on patient age (<or= 45 years or > 45 years) and sex also were performed.

RESULTS

Prior to urgent appendectomy, 18.5% of patients underwent preoperative CT in 1998 compared with 93.2% of patients in 2007. The negative appendectomy rate for women 45 years of age and younger decreased from 42.9% in 1998% to 7.1% in 2007. However, the timing of the decline in negative appendectomy rates for women 45 years and younger could not be proved to be associated with the increase in CT use. There was no significant trend toward a lower negative appendectomy rate for men regardless of age or for women older than 45 years of age with increased use of preoperative CT. The shift from single-detector CT to multidetector CT and the use of decreasing section thickness also correlated with a reduction in false-positive diagnoses.

CONCLUSION

Rising utilization of preoperative CT and advances in technology coincided with a decrease in the negative appendectomy rate for women 45 years and younger but not in men of any age or women older than 45 years.

Diagnostic accuracy of noncontrast computed tomography for appendicitis in adults: a systematic review

STUDY OBJECTIVE

We seek to determine the diagnostic test characteristics of noncontrast computed tomography (CT) for appendicitis in the adult emergency department (ED) population.

METHODS

We conducted a search of MEDLINE, EMBASE, the Cochrane Library, and the bibliographies of previous systematic reviews. Included studies assessed the diagnostic accuracy of noncontrast CT for acute appendicitis in adults by using the final diagnosis at surgery or follow-up at a minimum of 2 weeks as the reference standard. Studies were included only if the CT was completed using a multislice helical scanner. Two authors independently conducted the relevance screen of titles and abstracts, selected studies for the final inclusion, extracted data, and assessed study quality. Consensus was reached by conference, and any disagreements were adjudicated by a third reviewer. Unenhanced CT test performance was assessed with summary receiver operating characteristic curve analysis, with independently pooled sensitivity and specificity values across studies.

RESULTS

The search yielded 1,258 publications; 7 studies met the inclusion criteria and provided a sample of 1,060 patients. The included studies were of high methodological quality with respect to appropriate patient spectrum and reference standard. Our pooled estimates for sensitivity and specificity were 92.7% (95% confidence interval 89.5% to 95.0%) and 96.1% (95% confidence interval 94.2% to 97.5%), respectively; the positive likelihood ratio=24 and the negative likelihood ratio=0.08.

CONCLUSION

We found the diagnostic accuracy of noncontrast CT for the diagnosis of acute appendicitis in the adult population to be adequate for clinical decisionmaking in the ED setting.

ED endovaginal pelvic ultrasound in nonpregnant women with right lower quadrant pain

INTRODUCTION

We hypothesized that emergency physician-performed endovaginal ultrasound (EVUS) would change diagnostic decision making in nonpregnant women with right lower quadrant (RLQ) pain.

METHODS

A prospective cohort of female patients was enrolled at an urban emergency department (ED). Inclusion criteria were RLQ pain, hemodynamic stability, and a strong suspicion for appendicitis or right adnexal pathology. Treating physicians were queried regarding pre- and post-ED EVUS probability of disease, differential diagnoses, consultation, and management. Positive findings included large cysts or multitissue densities, tubal dilation, uterine enlargement/mass, and extensive peritoneal fluid.

RESULTS

With a positive ED EVUS, mean physician probability increased for gynecologic (24%) and decreased for both surgical (14%) and medical (20%) disease. With a negative ED EVUS, mean physician probability increased for surgical disease (5.3%) and decreased for gynecologic disease (18.6%).

CONCLUSION

Emergency department EVUS changes physician diagnostic decision making in nonpregnant women with undifferentiated RLQ pain.

Abdominal CT Scanning in Reproductive-Age Women with Right Lower Quadrant Abdominal Pain: Does Its Use Reduce Negative Appendectomy Rates and Healthcare Costs?

Although acute appendicitis is the most frequent cause of the acute abdomen in the United States, its accurate diagnosis in reproductive-age women remains difficult. Problems in making the diagnosis are evidenced by negative appendectomy rates in this group of 20 per cent to 45 per cent. Abdominal CT scanning has been used in diagnosing acute appendicitis, but its reliability and usefulness remains controversial. There is concern that the use of CT scanning to make this diagnosis leads to increased and unwarranted healthcare charges and costs. The purpose of our study is to determine if abdominal CT scanning is an effective test in making the diagnosis of acute appendicitis in reproductive-age women (age, 16–49 years) with right lower quadrant abdominal pain and to determine if its use is cost-effective. From January 2003 to December 2006, 439 patients were identified from our academic surgical database and confirmed by chart review as undergoing an appendectomy with a pre- or postoperative diagnosis of acute appendicitis. Data, including age, presence and results of preoperative abdominal CT scans, operative findings, and pathology reports were reviewed. Comparison of patients receiving a preoperative CT scan with those who did not was performed using chi-squared analysis. In the subgroup of reproductive-age women, there was a significant difference in negative appendectomy rates of 17 per cent in the group that received abdominal CT scans versus 42 per cent in the group that did not ( P < 0.038). After accounting for the patient and insurance company costs, abdominal CT scan savings averaged $1412 per patient. Abdominal CT scanning is a reliable, useful, and cost-effective test for evaluating right lower quadrant abdominal pain and making the diagnosis of acute appendicitis in reproductive-age women.

Current Concepts in Imaging of Appendicitis

Appendicitis is the most common cause of acute abdominal pain requiring surgery. Early diagnosis is crucial to the success of therapy. CT and ultrasound are widely recognized as very useful in the timely diagnosis of appendicitis. MR imaging is emerging as an alternative to CT in pregnant patients and in patients who have an allergy to iodinated contrast material. This article reviews the current imaging methods and diagnostic features of appendicitis.

Computed tomography-based clinical diagnostic pathway for acute appendicitis: prospective validation

BACKGROUND

Despite our regular use of CT for suspected appendicitis, a recent study at our institution demonstrated a negative appendectomy (NA) rate similar to our rate 15 years earlier. Based on analysis of this data, a diagnostic pathway was implemented prospectively for adult patients with possible appendicitis.

STUDY DESIGN

Rates of CT, NA, and perforation for nonpregnant patients over 14 years of age undergoing appendectomy between August 2001 and August 2002 (PRE) were compared with prospectively collected pathway data (PATH, August 2004 to August 2005). All PATH patients were evaluated by a surgeon before CT. All females underwent CT. CT was obtained in male patients with low suspicion or pain for longer than 48 hours. After negative CT, patients were discharged from the emergency department or admitted for observation.

RESULTS

There was a substantial decrease in NA rate after implementation of the pathway (4% PATH [8 of 183] versus 16% PRE [31 of 196], p < 0.001), without a change in the rate of perforation (11% PATH [20 of 175] versus 8% PRE [13 of 165], p = 0.28) or frequency of preoperative CT (59% PATH [108 of 183] versus 60% PRE [118 of 196], p = 0.84).

CONCLUSIONS

Frequent appendiceal CT alone does not ensure surgical diagnostic accuracy. CT need not be used in all patients to achieve very low NA rates. An evidence-based diagnostic algorithm incorporating early surgical evaluation, objective criteria for preoperative CT, deterrence of early operation after negative CT, and use of CT to facilitate safe discharge led to substantial improvements in the care of our adult population presenting with possible appendicitis.