Retrospective survey of practices of endoscopic retrograde cholangiopancreatography in Kenyatta National Hospital

¹Department of Clinical Medicine and Therapeutics, School of Medicine, University of Nairobi, P.O. Box 19676 - 00202, Nairobi, Kenya

^&Corresponding author
Sajni Pabari, Department of Clinical Medicine and Therapeutics, School of Medicine, University of Nairobi, P.O. Box 19676 - 00202, Nairobi, Kenya

Introduction    

Endoscopic retrograde cholangiopancreatography (ERCP) was introduced in 1968, initially as a diagnostic tool but has since evolved to be utilized mainly as a therapeutic intervention. Diagnosis has mainly been taken over by less invasive procedures such as abdominal ultrasound, computed tomography (CT) scan, endoscopic ultrasound (EUS), and magnetic resonance cholangiopancreatography (MRCP) [1]. Use of ERCPs in developing countries like our own had not started until the last 2 decades. It was first done in the Kenyatta National Hospital (KNH) Endoscopy Unit in 2014. Before then we relied on open surgery for the management of simple pancreato-biliary conditions that could easily be treated with the minimally invasive ERCP procedure, further increasing morbidity and mortality, especially in poor surgical candidates [2]. Currently, approximately 100 - 150 ERCPs are performed by both experienced and trainee gastroenterologists and surgeons at our endoscopy unit every year, making it the second most common gastroenterology procedure done at 16.1% according to a study done by Waweru et al. in KNH [3].

Methods     

Study design: this was a retrospective survey of ERCPs done at the KNH Endoscopy Unit between August 2014 and December 2019.

Primary objectives: a) to describe the indications, findings, and interventions performed during ERCP procedures done in the KNH Endoscopy Unit; b) to determine the incidence and type of complications of ERCP done in the KNH endoscopy Unit.

Secondary objective: to document the diagnostic imaging that was performed before the ERCP procedure.

Inclusion criteria: all files with ERCP procedure performed at the endoscopy unit in KNH.

Exclusion criteria: endoscopic retrograde cholangiopancreatography without an internet protocol (IP) number or incomplete IP numbers recorded in the endoscopy register.

Definition of terms

Complications: complications were unfavorable/unwanted outcomes of the ERCP procedure occurring during, immediately after, or up to 4 weeks post-ERCP.

i) Pancreatitis

Mild: clinical pancreatitis, amylase >3x the normal at 24 hours post-ERCP and requiring/extension of planned admission to 2 - 3 days.

Moderate: pancreatitis necessitating admission for 4 - 20 days.

Severe: hospital admission for >10 days, or, development of infection, pseudocyst, hemorrhagic pancreatitis, phlegmon, or requiring percutaneous drainage or surgery [4].

ii) Infection: the Tokyo guidelines 2007 were used to define ERCP-related infections [4].

Cholangitis: a) presence of fever >38°C, chills, abdominal pain, and jaundice, or b) laboratory evidence of biliary obstruction and inflammation, or c) imaging evidence of biliary obstruction, and d) no evidence of acute cholangitis in the one week before the ERCP procedure.

Cholecystitis: patient with any 1 of; murphy´s signs positive, a mass, pain, or tenderness over the right upper quadrant, and a) one of; fever, elevated white blood cells, elevated C-reactive protein.

Bacteremia: a) presence of fever, chill, and b) positive bacterial blood cultures post-ERCP.

iii) Hemorrhage: hemorrhage was defined as per cotton et al. as:

Mild: drop in Hb of <3 [g/dl] without need for transfusion

Moderate: four units of blood transfused without need of angiographic or surgical intervention.

Severe: 5 units of blood transfusion, or surgical/angiographic intervention needed [4].

iv) Perforation: perforation is the puncturing of the viscus by the duodenoscope or accessory equipment used during the ERCP procedure.

v) Cardiopulmonary complications

Hypotension: a drop in blood pressure at any time during the ERCP procedure below a systolic pressure of 90mmHg and diastolic pressure of 60mmHg.

Hypoxia: a drop in oxygen saturation below 90% at any point during the ERCP procedure.

Arrythmias: any new change in heart rate pattern at any point during the ERCP procedure.

Air embolism: unwanted entry of air into the vasculature during the ERCP procedure, suspected with sudden deterioration in vital signs.

Completion rate: this is the number of ERCPs that had successful interventional outcomes as intended for any given indication.

Recruitment: the investigator and research assistant retrieved the I.P numbers of all ERCPs done between August 2014 and December 2019 from the register in the Endoscopy Unit at KNH. With the help of records officers, we retrieved the files of the same patients. All patients were then coded with unique identification numbers to maintain patient confidentiality. Each ERCP procedure was then subcoded to enable identification of repeat procedures (≥ 2 ERCPs) done on the same patient. To ensure completeness in data collection all consecutive ERCPs performed at the KNH endoscopy unit found in the collected files were used for data collection whether recorded in the register or not. The investigators collected relevant socio-demographic and clinical-pathological data using the pre-specified study pro forma. Missing files were analyzed for demographic characteristics from the endoscopy register. Secondary sources of data collection such as the nursing cardex and anesthetist chart showing continuous oxygen saturations, blood pressure, and heart rate were sought for files with absent/incomplete procedure notes for completeness of data. All patients had a hemogram and coagulation profile performed before the procedure, and any complications warranted further investigations which were used for documentation of bleeding complications. All data collected was stored in a password-protected laptop computer which was in the sole custody of the primary investigator.

Data analysis: data was entered and managed in the Microsoft Access database. At the end of data entry, cleaning was performed and the data was exported into SPSS version 23.0 statistical software for analysis. Descriptive statistics was used for demographic and clinical characteristics. Endoscopic retrograde cholangiopancreatography indications, findings, interventions, and complications were summarized and presented as percentages. In addition, pre-procedural imaging done before the ERCP procedure was determined and presented as proportions of the studied population.

Ethical consideration: data collection and presentation were done with ethical approval from the Department of Clinical Medicine and Therapeutics of the University of Nairobi and the KNH research and ethics committee.

Results     

Six hundred and sixty-nine (669) ERCPs were registered in the endoscopy register in the aforementioned study period yielding 408 files. Two hundred and sixty-one (261) files were excluded leaving 542 ERCPs for analysis. All of them were performed as inpatients. Out of the 269 excluded files; 91 files were repeat entries of the same ERCP and hence could not be analyzed again; 38 files had wrong IP numbers, meaning that files with the recorded IP numbers did not match the name on the register and had no ERCP documented in them; 34 files had missing IP numbers and therefore they could not be retrieved and 98 files were missing from the records department.

Demographic data: the mean and median age were 55.9 (SD 16.5) and 56.0 (IQR 44.0-68.0) years respectively. Our male-to-female ratio was 1:1.5. Out of the 542 ERCPs recorded, 156 were referrals from various facilities, the majority coming in from within Nairobi. Over one-third of the patients were residents of Central Province and approximately 30% had comorbidities, the commonest being hypertension.

Anesthesia: eighty-two point three percent (82.3%) of ERCPs were done under anesthesia. The majority of the patients received 3 sedative drugs, most commonly propofol, midazolam, and fentanyl. Nine were under paralysis.

Procedure completion rate: seventy-eight-point nine percent (78.9%) of the ERCPs were completed. Endoscopic retrograde cholangiopancreatography that could not be done due to peri-procedural complications and failure to visualize papilla was not included in the analysis (Table 1).

Indications: there were 309 (63.6%) ERCPs with benign biliary indications and 177 (36.4%) ERCPs with malignant indications. The most common pancreatic indication was carcinoma of the head of the pancreas (Table 2).

Findings of endoscopic retrograde cholangiopancreatography: common bile duct (CBD) strictures were the most common finding at 44.5% (Table 3).

Endoscopic retrograde cholangiopancreatography intervention performed: as for the indications, the majority of the interventions performed were biliary rather than pancreatic. Fourty-one (41) ERCPs had a biopsy done and only 4 had brushing performed. Out of this, only 26 had histology reports in the files (Table 4, Table 5).

Complications: the overall complication rate was 36.0% (Table 6).

Discussion     

This case illustrates the diagnostic challenges posed by cardiac asthma, which can closely mimic bronchial asthma. Both conditions present with wheezing and dyspnea, necessitating a thorough clinical assessment and diagnostic workup to differentiate between them. Key distinguishing features include the presence of cardiac risk factors, physical examination findings suggestive of heart failure, and diagnostic tests such as echocardiography to assess cardiac function. Failure to recognize cardiac asthma can lead to delayed diagnosis, inappropriate treatment, and potentially adverse outcomes.

Research findings indicate that the prevalence of asthma among individuals aged 65 years and older falls between 6.5% and 10.4%, as reported in various studies, including the cardiovascular health study, where an 8% prevalence was noted [1]. Similarly, data from the National Health Interview Survey conducted by the National Center for Health Statistics, CDC, revealed a 7.5% prevalence in this age group [2]. However, within the elderly population diagnosed with congestive heart failure, approximately 35% exhibit wheezing consistent with cardiac asthma [3], surpassing the prevalence of asthma in this demographic by more than 3.5 times. This suggests that a significant proportion of asthma diagnoses in the elderly may represent cases of cardiac asthma.

Treatment guidelines typically advocate for vasodilators like nitrates or non-invasive ventilation, such as continuous positive airway pressure (CPAP), in managing cardiac asthma. However, access to such modalities may be limited in remote and rural areas. In such contexts, loop diuretics are often the mainstay, albeit suboptimal when used alone. In this case, a pragmatic approach was adopted, combining hydralazine, a vasodilator, with furosemide to enhance diuresis by optimizing renal blood flow. Additionally, the patient received losartan and spironolactone for heart failure management. This comprehensive regimen led to significant improvement, as evidenced by the resolution of wheezing and successful recovery.

Conclusion     

Cardiac asthma represents a diagnostic dilemma, often masquerading as bronchial asthma. Clinicians must maintain a high index of suspicion for cardiac etiologies, especially in patients with known cardiovascular risk factors or atypical asthma presentations. Timely recognition and appropriate management are paramount for optimizing patient care and preventing unnecessary misdiagnosis-related morbidity and mortality. This case underscores the importance of a comprehensive approach to respiratory symptoms, integrating clinical judgment, diagnostic testing, and therapeutic interventions tailored to the underlying etiology.

Research limitations: the record-keeping in the files is hampered with the completion of data collection. We pursued secondary sources to ensure completeness.

Research strength: a large number of ERCPs that were unselected were analyzed, giving power to the conclusions drawn from the study.

Recommendations: improved training methods such as simulation training should be employed to increase the completion rate of ERCPs. Further studies need to be done to ascertain the cause of increased cardiopulmonary complications. Cytology brushes need to be procured and operationalists should use standard biopsy forceps when brushes are unavailable to enable optimal utilization of ERCP. We also need to obtain cholangioscopy to aid in direct biopsies as majority of our indications were malignant. Improved documentation, record keeping and employing a computerized system will aid future data collection. Regular surveys for the purpose of quality improvement.

Competing interests     

The authors declare no competing interests.

Authors' contributions     

All the authors have read and agreed to the final manuscript.

Tables     

Table 1: reason for incomplete endoscopic retrograde cholangiopancreatography procedures

Table 2: indications of endoscopic retrograde cholangiopancreatography

Table 3: findings of endoscopic retrograde cholangiopancreatography

Table 4: biliary interventions performed during endoscopic retrograde cholangiopancreatography procedure

Table 5: pancreatic interventions performed during endoscopic retrograde cholangiopancreatography procedure

Table 6: complications recorded for endoscopic retrograde cholangiopancreatography procedures

References