Preoperative Factors Associated with Appendiceal Tumors in Nonelective Appendectomy

Introduction

Appendectomy is one of the most commonly performed general surgery operations, and the incidence of acute appendicitis is reported in more than 250,000 cases per year in the United States.^1–3 The rates of incidental malignant appendiceal tumors diagnosed on final pathology following appendectomy for presumed appendicitis in the United States is <2%.^1–3 Appendiceal tumors commonly present as acute appendicitis in more than 50% of the cases, and the majority of these diagnoses are made postoperatively on the final pathology reports.^1–5 Although most patients in this cohort have early-stage cancers, rendering appendectomy sufficient for treatment, a small percentage of these patients may ultimately require further oncologic resection.^1–3

Some studies have investigated risk factors for appendiceal tumors, specifically for appendiceal carcinoid. Unfortunately, these studies have reported conflicting results, citing younger versus older age as a risk factor.^1–3 Furthermore, these studies are limited by small sample size in single-institution studies, due to the rarity of this tumor type.^1–3 No large studies have investigated the preoperative factors, including imaging findings associated with incidental appendiceal neoplasms. Our study aims to identify factors, including patient demographics, laboratory results, and imaging results, within the American College of Surgeons National Surgical Quality Improvement Program (ACS-NSQIP) database that are associated with an increased risk of incidental appendiceal tumors that present as appendicitis.

Materials and Methods

Using the targeted appendectomy ACS-NSQIP database, we identified all patients who underwent nonelective appendectomy for acute appendicitis in 2016, provided by the ACS-NSQIP Procedure Targeted Participant Use Data File (PUF). Their files were merged, and more specific appendectomy-related variables were investigated within the targeted database. Final pathology was reported as “consistent with appendicitis,” “not consistent with appendicitis,” “other appendiceal pathology,” “pathology report not available,” “result is indeterminate or uncertain,” “tumor involving appendix,” or “unknown.” Due to the limited numbers of appendiceal tumor cases, patients with tumors were compared with randomly selected patients with acute appendicitis pathology using a 1:4 case-controlled cohort, respectively. The case-controlled model ensured adequate power for analyses, and the two cohorts were not matched for any variables.

Preoperative patient and diagnostic variables were investigated using univariate analysis to determine factors associated with tumor pathology. The variables included were preoperative patient demographics, laboratory results, and different preoperative imaging types such as computed tomography (CT) and ultrasound (US), or no preoperative imaging. The formal radiology reads were classified as “indeterminate,” “not consistent with appendicitis,” “consistent with appendicitis,” and “those without imaging.” Variables found to be statistically significant in univariate analysis were then included in a stepwise logistic regression analysis comparing those with tumor pathology with those with acute appendicitis pathology. A backward-elimination approach was used and predictors with P > .10 were dropped at each step. All statistical analyses were performed using SAS version 9.3 (SAS Institute, Cary, NC) and P < .05 was considered statistically significant.

Results

A total of 12,376 patients underwent nonelective appendectomy in the 2016 ACS-NSQIP database; 127 (1.02%) patients had appendiceal tumor pathology. After using the 1:4 randomly selected case-controlled cohort for appendiceal tumor to acute appendicitis pathology, 127 had tumor pathology and 508 patients had pathology consistent with appendicitis. On univariate analysis, factors including female gender, age, American Society of Anesthesiologist (ASA) classification, dyspnea, nonindependent functional status, lower preoperative albumin, and lower preoperative white blood cell (WBC) count were statistically significantly associated with tumor pathology (Table 1). Our analysis showed no significant difference in patient presentations such as perioperative appendicitis perforation, abscess formation, or changes in postoperative complications in the appendiceal tumor versus acute appendicitis group (Table 2). Operative time in the appendiceal tumor cohort was significantly higher than that in the appendicitis cohort (average 65.2 ± 73.5 versus 53.1 ± 29.2 minutes, Table 2).

Following multivariate analysis, female gender, increased age, and lower preoperative WBC count were statistically significant predictors for final tumor pathology (Table 3). Females consisted of 43.9% of patients in the appendicitis pathology group versus 59.8% in the group with tumor pathology (Table 1). The odds of having tumor pathology were 82% higher for females than for males (Table 3). The average age was 39.7 ± 17.1 years for patients with normal appendicitis pathology, while the group with tumor pathology had an average age of 45.2 ± 18.5 years (Table 1). Interestingly, for every 1-year increase in age, there was an increased risk of having incidental appendiceal tumor from emergent appendectomy by 2% (Table 3). Finally, the preoperative WBC count was on average elevated (13.2 ± 4.3 thousand cells per cm³) in patients who had acute appendicitis pathology, while the average WBC count was slightly above normal limits at 11.2 ± 4.3 thousand cells per cm³ (Table 1). This difference was significant. Comparatively, the odds of having tumor pathology increased by 3% for each one-unit decrease in WBC count (Table 3).

Preoperative imaging data analysis revealed that clinicians used CT scans as the predominant imaging modality for diagnosing acute appendicitis, followed by US (Table 1). A total of 3.62% of patients did not have any preoperative imaging (Table 1). After multivariate analysis, the odds of having tumor pathology in patients with radiology reads of “indeterminate” and “not consistent with appendicitis” performed were 2.38, 3.29, and 3.17 times higher than patients who had radiology reading of “consistent with appendicitis,” respectively (Table 4). Regardless of imaging modalities, pathology results sorted by imaging reports showed a significantly higher percentage of patients with appendicitis pathology with imaging read as “consistent with appendicitis,” while those with tumor pathology had significantly higher percentage of imaging read as “not consistent with appendicitis,” “indeterminate,” or no preoperative imaging (Table 5).

Discussion

Appendiceal tumors are rare and are estimated to occur in <2% of patients who undergo appendectomy for acute appendicitis.^1–3 Based on ∼250,000 appendectomies performed annually for acute appendicitis in the United States, 2000–5000 patients will have an incidental appendiceal tumor discovered on final pathology.^1–6 As more than 50% of appendiceal tumors present as acute appendicitis, these tumors are most often diagnosed incidentally on pathology reports following appendectomy.^1–7 Appendectomy alone is sufficient surgical treatment for up to 70% of the cases, however, the remaining patients may require additional surgical resection following appendectomy.^1–4,6 There are no established risk factors for appendiceal cancer, particularly in those patients presenting with acute appendicitis, nor are there guidelines for the management of suspected appendiceal tumor with the diagnosis of acute appendicitis. ⁶ Our study is the first large-scale study that attempts to identify patient and imaging risk factors associated with incidental appendiceal neoplasms that present as appendicitis using the targeted ACS-NSQIP appendectomy.

Our study reveals a demographic of patients with identifiable preoperative factors (age, gender, and WBC) that are associated with an increased risk of appendiceal tumor. In addition, the study also showed a significant increase in operative time in the tumor cohort compared with appendicitis. Possible reasons for this include increased technical difficulty and increased surgical decision-making time when the surgeon has a high suspicion of appendiceal tumor.

While the advanced age and lower WBC are identified as statistically significant factors in our analysis and other studies, these differences may not be appreciable or relevant in the clinical setting. Clinically, it is not surprising that patients with appendiceal tumors who present as acute appendicitis tend to be more advanced in age. The majority of primary appendiceal tumors are neuroendocrine tumors (NETs) and epithelial neoplasms that present in the fourth and fifth decades of life, while patients with acute appendicitis usually present during the first three decades in life.^6–10 Patients with occult appendiceal tumors also tend to have lower WBC counts as the tumor causes local inflammation versus the systemic response, which is common in acute appendicitis. ¹⁰ Finally, our data demonstrate increased risk of occult appendiceal tumor in females who present with acute appendicitis, which is similar to other studies that show a slight female predominance in appendiceal carcinoid tumor, but not in other types of appendiceal cancer.^7–10 Surgeons should have a heightened awareness for these preoperative factors and consider intraoperative frozen sections when they are present. Preoperative conversations should mention the potential need for reoperation in the event additional resection (hemicolectomy) is indicated.

In addition to patient demographics and preoperative laboratory findings, patients with preoperative imaging results that were “indeterminate” or “not consistent with appendicitis” have significantly increased odds for appendiceal tumor pathology in our study, regardless of imaging modality. Teixeira et al. showed that having an inflammatory appendiceal phlegmon on imaging was associated with an increased risk of occult appendiceal tumor. ¹¹ Other studies have identified more specific radiographic features such as a small localized mass at the appendiceal tip in NETs, mucin-distended appendix with mucinous appendiceal neoplasm, soft tissue mass in nonmucinous adenocarcinoma, and diffuse wall thickening with aneurysmal dilation in primary appendiceal lymphoma.^10,12,13 While specific detailed characteristics of the images read as “indeterminate” or “not consistent with appendicitis” are not available within the ACS-NSQIP-targeted database, we surmise that there may be common findings among these images that may increase suspicion for occult tumor.

A few single-institution studies identified risk factors for having occult appendiceal tumor in acute appendicitis with conflicting results. Schwartz et al. reported that younger patients have a significantly higher risk of occult appendiceal carcinoid neoplasm than older patients, ² while Van Gompel et al. reported that females (70% of carcinoid versus 45% of noncarcinoid patients), older age (average 41 years old versus 30 years old), and having a lower WBC count (average 10.8 versus 14.2 × 1000/mL) were predictors of having incidental carcinoid appendiceal tumor on final pathology for acute appendicitis. ³ Interestingly, our data identified female gender, increased age, and lower preoperative WBC count as risk factors for occult appendiceal tumors, consistent with the results of Van Gompel et al. ³

A few studies attempt to address recommendations for the perioperative management of potential occult appendiceal tumors. For example, if there is a suspicious mass or other abnormalities involving more than the appendix on imaging, Leonards et al. recommend that patients should be treated with antibiotics initially and undergo colonoscopy to exclude possible synchronous colonic cancer. ¹⁰ In contrast, some studies suggest intraoperative frozen sections if there is a suspicion of appendiceal tumor intraoperatively, to make the determination for right hemicolectomy versus simple appendectomy intraoperatively.^{4,8,11,13–15}

The need for additional resection affects up to 20% of patients who had incomplete oncologic resections of incidental appendiceal tumors discovered after simple appendectomy. ⁴ Despite this, there is a lack of studies on patients who require return to the operating room for complete oncologic resection after initial appendectomy, likely due to its rarity. One might expect patients who require further resection to have increased risk of perioperative morbidity in a reoperative field in addition to delay of potential chemotherapy. However, there are no published data to investigate whether patients who require further operation have increased mortality or if initial nonoperative management of acute appendicitis for cases with high suspicion for occult appendiceal tumor should be considered.^10,13–16 These are areas that will require further investigation.

Our study has several limitations. It is a retrospective study based on the ACS-NSQIP database that does not include detailed information regarding abdominal pain on presentation, tumor pathology involving the base versus tip of the appendix, various tumor histologies, or details that differentiate between the imaging results of those with “indeterminate” and “not consistent with appendicitis” radiology interpretations. In addition, the database does not report if patients have tumor pathology along with histologic signs of appendicitis. For example, patients with tumor pathology may develop appendicitis secondary to local tumor inflammation and stasis.^6–8 The database does not allow us to differentiate such details nor for further inquiry into patients who needed additional oncologic resection after initial appendectomy. Lastly, due to the limited numbers of appendiceal tumor cases, we had to adjust our analysis with a nested case/control study to ensure adequate power for statistical analysis.

Conclusion

Incidental appendiceal tumors are a rare cause of acute appendicitis. These tumors are difficult to diagnose before definitive pathologic analysis due to a similar clinical presentation as patients with acute appendicitis. To reduce the risk of additional resection after initial appendectomy, clinicians should have a high suspicion for occult appendiceal tumor in patients who are female, older in age, with a borderline elevated preoperative WBC count, and imaging results of “indeterminate” or “not consistent with appendicitis.” Clinicians should have a thorough preoperative conversation, with patients meeting these criteria, that includes the potential need for additional resection. Intraoperative frozen sections may be a useful tool to help guide surgical management.