Association of Colonoscopy with Risk of Appendicitis

Letters

Author Contributions: Drs Merola and Pei had full access to all the data in the rent Procedural Terminology (CPT) code between January 2009 study and take responsibility for the integrity of the data and the accuracy of and June 2014, excluding sigmoidoscopy or incomplete colo- the data analysis. Study concept and design: Merola, Resio, Davis, Pei. noscopy codes. We sought over the following year a CPT code Acquisition, analysis, or interpretation of data: All authors. for appendectomy, an International Classification of Diseases, Drafting of the manuscript: Merola, Arnold, Ibarra, Pei. Ninth Revision (ICD-9) code for appendicitis, both CPT and Critical revision of the manuscript for important intellectual content: Merola, ICD-9 codes, or the appendectomy CPT code with a discharge Arnold, Luks, Resio, Davis, Pei. Statistical analysis: Merola, Arnold, Luks. diagnosis of appendicitis. We calculated the incidence rate ra- Administrative, technical, or material support: Resio. tio (IRR) for appendicitis within 1 week after undergoing a colo- Study supervision: Davis, Pei. noscopy and for the following 51 weeks. χ2 Statistics and 95% Conflict of Interest Disclosures: None reported. confidence intervals were calculated. Meeting Presentation: This paper was presented at the Annual Meeting of the Association of VA Surgeons; May 7, 2017; Houston, Texas. Results | Appendicitis or an appendectomy were coded more 1. Kyzer S, Gordon PH. The prophylactic use of ureteral catheters during frequently 1 week after a colonoscopy than during the follow- colorectal operations. Am Surg. 1994;60(3):212-216. ing 51 weeks (Table). Age subgrouping did not change this, and 2. Bothwell WN, Bleicher RJ, Dent TL. Prophylactic ureteral catheterization in colon surgery: a five-year review. Dis Colon Rectum. 1994;37(4):330-334. appendicitis was not increased in weeks 2 to 4 (data not shown). By contrast, appendectomy was not increased 1 week after a 3. Chahin F, Dwivedi AJ, Paramesh A, et al. The implications of lighted ureteral stenting in laparoscopic colectomy. JSLS. 2002;6(1):49-52. bronchoscopy, knee replacement procedure, cataract sur- 4. Nam YS, Wexner SD. Clinical value of prophylactic ureteral stent indwelling gery, or knee arthroscopy in the cohort (Table). during laparoscopic colorectal surgery. J Korean Med Sci. 2002;17(5):633-635. Recognizing the challenges of administrative data, we re- 5. St-Louis E, Iqbal S, Feldman LS, et al. Using the age-adjusted Charlson viewed records from each patient that were CPT-coded for ap- comorbidity index to predict outcomes in emergency general surgery. J Trauma pendectomy. Eliminating patients with appendiceal neo- Acute Care Surg. 2015;78(2):318-323. plasms, incidental appendectomies, or no appendectomy in 6. da Silva G, Boutros M, Wexner SD. Role of prophylactic ureteric stents in the operative note, we included only patients with surgical and colorectal surgery. Asian J Endosc Surg. 2012;5(3):105-110. histologic findings that were consistent with appendicitis ex- cept for 2 patients outside the first week with clear operative Association of Colonoscopy With Risk of Appendicitis descriptions of acute appendicitis but unavailable pathology Appendicitis can occur after a patient undergoes a colonoscopy,1 reports. Only 12 patients (74%) who were administratively iden- but because both are common, it is unclear whether a colonos- tified as having undergone an appendectomy actually devel- copy increases appendicitis risk. We hypothesized that pa- oped appendicitis 1 week after undergoing a colonoscopy. None tients develop appendicitis more frequently within 1 week af- had appendiceal or cecal biopsies, appendicoliths, or perfo- ter undergoing a colonoscopy than during the following 51 rated appendicitis. Seventy-nine administratively identified weeks. patients (77%) actually had appendicitis over the following 51 weeks. Considering only validated patients from the CPT- Methods | After obtaining approval and a waiver of informed based strategy yielded an IRR of 6.8 (95% CI, 3.4-12.6) for ap- consent from the University of North Dakota and Fargo Vet- pendicitis within a week after colonoscopy (P < .001). Five pa- erans Affairs Medical Center institutional review boards, we tients (42%) with appendicitis in week 1 had preexisting identified 392 485 veterans from US Department of Veterans symptoms that were investigated by a colonoscopy. Even ex- Affairs administrative data with a screening colonoscopy Cur- cluding these yielded an IRR of 4.5 (95% CI, 1.8-9.8; P < .001).

Table. Incidence of Appendicitis or Appendectomy Among Veterans Undergoing a Prior Procedure

No. of Patients Incidence Undergoing Prior Weeks Rate Ratio P Value Analysis Strategy Prior Procedure Procedure Week 1 2-52 (95% CI) by χ2 Appendicitis ICD-9 codes Colonoscopy 392 485 26 175 7.6 (4.8-11.5) <.001 Appendectomy CPT codes Colonoscopy 392 485 19 112 8.7 (5.1-14.2) <.001 Both ICD-9 and CPT codes Colonoscopy 392 485 12 82 7.5 (3.7-13.8) <.001 Appendectomy CPT codes with Colonoscopy 392 485 10 79 5.7 (2.6-11.0) <.001 appendectomy discharge diagnosis Appendectomy CPT codes Bronchoscopy 16 761 1 9 5.7 (0.13-41.1) .09 Appendectomy CPT codes Total knee replacement 19 669 0 12 2.1 (0-22.7) .68 Appendectomy CPT codes Cataract 96 393 0 49 1.1 (.5-2.1) .57 Appendectomy CPT codes Knee arthroscopy 14 272 0 14 1.8 (0-19.1) .76 Appendectomy CPT codes with Colonoscopy 392 485 12 79 6.8 (3.4-12.6) <.001 chart validation Appendectomy CPT codes with Colonoscopy 392 485 7 79 4.5 (1.8-9.8) <.001 medical record validation and deletion of simultaneous symptoms

Abbreviations: CPT, Current Procedural Terminology; ICD-9, International Classification of Diseases, Ninth Revision.

90 JAMA Surgery January 2018 Volume 153, Number 1 (Reprinted) jamasurgery.com

Downloaded From: https://jamanetwork.com/ on 10/01/2021 Letters

Discussion | Although “healthy user biases” can select individu- 2. Shrank WH, Patrick AR, Brookhart MA. Healthy user and related biases in als who are more likely to use subsequent health care for other observational studies of preventive interventions: a primer for physicians. J Gen Intern Med. 2011;26(5):546-550. reasons,2 these biases seem unlikely to explain our results be- 3. Marquez Azalgara V, Sewitch MJ, Joseph L, Barkun AN. Rates of minor cause we compared appendicitis 1 week after a colonoscopy adverse events and health resource utilization postcolonoscopy. Can J with subsequent appendicitis in the same individuals. Nor is Gastroenterol Hepatol. 2014;28(11):595-599. appendicitis a diagnosis like hypertension that would be more 4. Decadt B, Sussman L, Lewis MP, et al. Randomized clinical trial of early likely to be an incidental finding on a precolonoscopy history laparoscopy in the management of acute non-specific abdominal pain. Br J Surg. and physical examination. Patients seek medical attention 1999;86(11):1383-1386. more frequently after undergoing a colonoscopy for colonos- 5. Baudet JS, Castro V, Redondo I. Recurrent ischemic colitis induced by colonoscopy bowel lavage. Am J Gastroenterol. 2010;105(3):700-701. copy-related complaints including bleeding, bloating, and 6. Drago L, Toscano M, De Grandi R, Casini V, Pace F. Persisting changes of pain,3 so one could hypothesize that an increased likelihood intestinal microbiota after bowel lavage and colonoscopy. Eur J Gastroenterol of complaining of abdominal symptoms might increase the Hepatol. 2016;28(5):532-537. likelihood of diagnosing appendicitis, but all of the index cases of appendicitis in the final analysis were pathologically con- firmed, so these were not false-positive diagnoses. Appendi- COMMENT & RESPONSE citis may sometimes resolve spontaneously, but the rate at which this occurs is difficult to distinguish from the resolu- Clinicopathological Factors Associated With tion of abdominal pain during conservative management that Prognosis of Patients With Intrahepatic was not appendicitis. Older studies that reported frequent non- Cholangiocarcinoma After Hepatectomy surgical resolutions used antibiotics that effectively treat much To the Editor We read with interest the article by Raoof et al1 appendicitis.4 evaluating many clinicopathologic variables possibly associ- While the actual IRR may differ and absolute risk is low, ated with postoperative outcomes for intrahepatic cholangio- these results suggest that undergoing a colonoscopy predis- carcinoma (ICC). The authors proposed a new prognostic score poses patients to appendicitis within 1 week. The mechanism for patients with ICC after hepatectomy. As their conclusions of this effect awaits elucidation. One patient developed symp- are different from previous studies, this may result in contro- toms after cleanout before the procedure. Some bowel prepa- versy about management of these patients. Therefore, this is- rations can precipitate ischemia5 or alter the microbiome,6 so sue deserves further discussion. the effect of bowel preparation also warrants exploration. First, the authors indicated that vascular invasion is not a Further mechanistic studies may identify patients who are at poor prognostic factor for patients with ICC, which is con- higher risk or elements of the bowel preparation or proce- trary to previous studies.2 Moreover, the prognostic nomo- dure that could be changed to decrease the risk. In the in- gram constructed by Yeh et al3 has been validated by 3 exter- terim, however, these results suggest that there is increased nal populations. We considered that there were some potential concern for the development of appendicitis among patients confounders affecting the authors’ conclusion. We noted that with persistent right lower quadrant pain after undergoing a there were 53 patients who underwent adjuvant chemo- colonoscopy. therapy and 20 patients with lymph node–positive metasta- ses in a subgroup of patients with tumors that were intrahe- Marc D. Basson, MD, PhD, MBA patic, small (<7 cm), and solitary (ISS). This means that most Daniel Persinger, MD of the other 119 patients without lymph node–positive metas- William P. Newman, MD tases had disease belonging to stage I according to the Ameri- can Joint Committee on Cancer TNM staging systems. How- Author Affiliations: Department of Surgery, University of North Dakota School of Medicine and Health Sciences, Grand Forks (Basson, Persinger); Fargo ever, adjuvant chemotherapy not only failed to improve survival Veterans Affairs Medical Center, Fargo, North Dakota (Basson, Persinger, but might even have promoted tumor recurrence for patients Newman); Department of Internal Medicine, University of North Dakota School with ICC at stage I.4 Therefore, adjuvant chemotherapy may of Medicine and Health Sciences, Grand Forks (Newman). counteract the relatively better prognosis of patients with sig- Corresponding Author: Marc D. Basson, MD, PhD, MBA, University of North nificantly less vascular invasion in the ISS subgroup. Dakota School of Medicine & Health Sciences, 1301 N Columbia Rd, Stop 9037, Grand Forks, ND 58202 ([email protected]). Second, there were 14 patients with ablation in both the Published Online: October 4, 2017. doi:10.1001/jamasurg.2017.3790 ISS and extrahepatic extension, large (>7 cm), and multifocal Author Contributions: Drs Basson and Newman had full access to all the data (ELM) subgroups. We do not consider that ablation belongs to in the study and take responsibility for the integrity of the data and the accuracy the category of surgery. Therefore, it is not appropriate that of the data analysis. the ablation is included in the extent of hepatectomy. More- Concept and design: Basson, Persinger. Acquisition, analysis, or interpretation of data: All authors. over, ablation is most efficacious in treating smaller tumors (≤3 Drafting of the manuscript: Basson, Persinger. cm), and it is difficult to achieve complete ablation when the Critical revision of the manuscript for important intellectual content: All authors. tumor size exceeds 5 cm.5 We surmise that ablation may pro- Statistical analysis: Basson, Newman. vide more significant local tumor control for patients in the ISS Administrative, technical, or material support: Persinger, Newman. Supervision: Basson. subgroup than those in ELM subgroup. This may potentially Conflict of Interest Disclosures: None reported. influence the authors’ result. 1. Shaw D, Gallardo G, Basson MD. Post-colonoscopy appendicitis: a case report Third, resection margin status was not mentioned in the and systematic review. World J Gastrointest Surg. 2013;5(10):259-263. study, and a previous study3 demonstrated that negative

jamasurgery.com (Reprinted) JAMA Surgery January 2018 Volume 153, Number 1 91

Downloaded From: https://jamanetwork.com/ on 10/01/2021