Splenic Injury in Outpatient Colonoscopy

Incidence, Risk Factors, Diagnosis, and Treatment

Since the initial description by Wherry and Zehner in 1974, numerous single-case reports or small case series of splenic injury caused by colonoscopies have been published—altogether 172 cases up to 2016 (1). Initial studies showed that the risk of splenic injury in inpatient colonoscopies is higher than assumed to date, at 1 case per 6000 colonoscopies (2). Colonoscopy, however, is an outpatient domain. According to data from the Techniker Krankenkasse (TK) health insurance fund, it is carried out as an outpatient procedure in 88% of cases. The small number of studies of complications after outpatient colonoscopies contributed little regarding splenic lesions (different objectives: complication rates under propofol sedation or the role of the training stage of the endoscopist) (3, 4). This study aimed to evaluate the incidence, course, diagnostic evaluation, therapy, and outcome of splenic lesions on the basis of 1.5 million outpatient colonoscopies.

Methods

The methods have already been described for inpatient colonoscopies (2). In sum, we identified from cross-sectoral routine data of the TK health insurance scheme (some 11 million TK insurance members) those individuals in whom complete outpatient colonoscopies were carried out in the time period from January 2016 through December 2020 according to the Gebührenordnungspositionen des Einheitlichen Bewertungsmaßstabs (EBM-GOP, the codes for physician reimbursement) and in which over time (≤7 days), indicators of a splenic injury were documented. Information related to the affected patients from difference service sectors were considered, codes of the International Classification of Diseases and Related Health Problems (ICD-10), EBM and OPS codes (the latter equivalent to the International Classification of Procedures in Medicine—ICPM).

All identified case constellations subsequently underwent a systematic single-case analysis by a gastroenterologist (HK) in coordination with a surgeon (TS) in order to verify or falsify temporal and causal associations between colonoscopy and splenic injury in the billing data—that is, the clinical course was reconstructed on the basis of ICD, EBM, and OPS codes. In patients undergoing abdominal (especially colonic) surgery, splenic injuries were categorized as complications of the surgery rather than those of colonoscopy (5). For this reason, only those splenic injuries were assigned to the colonoscopy where no indication of any other cause existed.

Results

A total of 1 443 341 complete colonoscopies (24% of which were screening colonoscopies) were analyzed in the study period. In temporal association with a colonoscopy, an ICD-10 code for splenic injury and/or an OPS code for a splenic intervention was found in 632 cases (0.044%), of which at an interval of 0–7 days after the colonoscopy in 163 an OPS code for imaging diagnostics and/or a splenic intervention. In 110 patients, no association with the colonoscopy existed (e.g. iatrogenic splenic injury in abdominal surgery, trauma, splenectomies for other indications). 7 of 8 persons with colonoscopy and abdominal sonography on the same day were not considered because of lacking plausibility.

In 46 patients the probability of splenic injury as a result of the colonoscopy was very high. In 1.4 million colonoscopies this corresponds to an incidence of 3.19 splenic lesions per 100 000 colonoscopies or one splenic injury per 31 400 colonoscopies. Women were more often affected (1 case per 24 848 endoscopies) (RR1.71) than men (1 case per 42 515 endoscopies). The age difference between the affected patients and the total population was small (Table).

Table

Demographic data of the original population with complete colonoscopy and the patients with complete colonoscopy and colonoscopy induced splenic injury

In 31 of the 46 cases (63%) the splenic lesion was diagnosed within 48 hours after the colonoscopy (Figure). In the cases with a delayed diagnosis, most were less severe splenic injuries according to the ICD codes, which were in part treated conservatively. In the initial phase, the main diagnostic mode was computed tomography, whereas in the later course it was sonography and magnetic resonance imaging (MRI) (Figure).

Figure

Time course of the diagnosis

Intensive care treatment and the need for transfusions (in 94% and 78% of operated patients, respectively) were predictors for an indication for surgery (n=19). Mostly—except for 2 cases with spleen-sparing surgery—splenectomies were carried out, in 17 of these 19 cases (89%) within 48 hours after the colonoscopy. No patient died.

Discussion

Analyses based on billing data can reflect healthcare provision in a complementary way to clinical and study data and capture data on rare complications. However, first it is necessary for an individual case analysis based on the ICD-10 and OPS codes to reconstruct the clinical course. In this way, causes other than colonoscopy for splenic injury or surgery (for example, abdominal surgeries, traumatic injuries) can be identified and ruled out (2). The strength of this study is the large number of cases of more than 1.4 million complete colonoscopies carried out on an outpatient basis. As the TK database reflects treatments in 13% of the German population, the results are likely to be representative for Germany as a whole.

A limitation lies in the natural drawbacks of billing data due to occasional incorrect coding and furthermore due to OPS and EBM codes which are solely day-specific. In spite of this, it was largely possible to reconstruct the treatment course exactly. The onset of symptoms remains unknown. Therefore, the timing of the imaging was used for the interval between colonoscopy and diagnosis of the splenic lesion (2).

The data show that splenic lesions mostly manifest within 48 hours but cases requiring surgery may still occur later. For reasons unknown to date—and in analogy to other analyses—women were mainly affected (1, 2). It is questionable whether age is an independent risk factor.

Splenic injuries are rarer in outpatient than in inpatient colonoscopies. Outpatients are probably healthier than the often multimorbid inpatients. Other risk factors under discussion—such as prior abdominal surgery, difficult passage through the colon, or strong pressure from the outside have not been confirmed, whereas Laanani (4) emphasizes the expertise of the endoscopist. It is possible that the strict quality assurance in outpatient colonoscopies are a contributing factor to decline in splenic lesions. With 2 million outpatient colonoscopies carried out every year in Germany, doctors need to consider this rare, potentially life-threatening complication (caution: Kehr’s sign) after colonoscopy and if needed immediately initiate sonographic or—better—computed tomography to make a diagnosis.

Herbert Koop, Christoph Skupnik, Torsten Schnoor,Dirk Horenkamp-Sonntag

Conflict of interest statement
The authors declare that no conflict of interest exists.

Manuscript received on 30 August 2024, revised version accepted on 8 January 2025

Translated from the original German by Birte Twisselmann, PhD.

Cite this as:
Koop H, Skupnik C, Schnoor T, Horenkamp-Sonntag D: Splenic injury in outpatient colonoscopy: Incidence, risk factors, diagnosis, and treatment. Dtsch Arztebl Int 2025; 122: 247–8. DOI: 10.3238/arztebl.m2025.0005

References

1. Jehangir A, Poudel DR, Masand-Rai A, Donato A: A systematic review of splenic injuries during colonoscopy: Evolving trends in presentation and management. Intern J Surg 2016; 33: 55–59

2. Koop H, Skupnik C, Schnoor T, Horenkamp-Sonntag D: Splenic injury ­associated with colonoscopy in hospitalized patients: Incidence, risk factors, management, and outcome. Endosc Intern Open 2024; 12: E1453−7

3. Bielawska B, Hookey LC, Sutradhar R, et al: Anesthesia assistance in outpatient colonoscopy and risk of aspiration pneumonia, bowel ­perforation and splenic injury. Gastroenterology 2018; 154: 77–85

4. Laanani M, Coste J, Blotière PO, Carbonnel F, Weill A: Patient, ­procedure and endoscopist risk factor for perforation, bleeding, and ­splenic injury after colonoscopies. Clin Gastroenterol Hepatol 2019; 17: 719–27

5. Forsberg A, Hammar U, Ekbom A, Hultcrantz R: A register-based study: Adverse events of colonoscopies in Sweden 2001–2013. Scand J ­Gastroenterol 2017; 52:1042–7