Differences in treatment strategies in the management of acute appendicitis in a county hospital in Guatemala and an academic teaching institution in the United States

Introduction

An appendicectomy is one of the most common urgent abdominal operations performed in the United States (US) and worldwide. ¹ However, the diagnosis of acute appendicitis and its management vary widely, particularly in rich compared to poor nations. The laparoscopic approach has become the dominant strategy in the USA, supported by a lower surgical site infection rate and reduced hospital stay compared to open appendicectomy. ² Likewise, NOM has emerged as a potential treatment of acute appendicitis in suitable patient populations with good outcomes. ³ This strategy has not been widely adopted in low-income countries but also within certain centres in the USA. ⁴

Identifying differences in management strategies in different countries might lead to an improvement in outcomes, though it must not be assumed that this is a one-way flow. HNSB is one of the four public hospitals in northern Guatemala and is the largest referral hospital in that region. It does not possess a CT scanner and only begun limited laparoscopic capability in 2013. ⁵ The most common urgent operation there is an appendicectomy. ⁶ Our study was undertaken to contrast differences in treatment stratagems between VA and HNSB and assess outcomes.

Methods

A retrospective analysis of adult (≥18-years-old) patients undergoing emergency appendicectomy between 2017 and 2019 was undertaken at HNSB. As there is no electronic medical record system here, data were extracted via paper chart review, which was not consistent in reporting all cases. Thus, a few paper charts (n = 27) that did not contain information on the age of the patient or the specific type of operation were excluded from the analysis. These data were compared to a previously published database at VA from 2006 to 2017, namely the VA North Texas Health Care System (VANTXHCS).^4,7 Data were extracted by a University Medical Student (MR) under the supervision of staff physicians at HNSB (AOH and CO) and at VANTXHCS (SH). Approval was obtained from the Institutional Review Board at VANTXHCS for veterans. Permission for the study was obtained by a written letter from the director of HNSB (CO) and its chief of surgery (AOH).

Data from the two patient populations were merged based on common variables available to both cohorts after extraction, though height was not available for patients from HNSB, so BMI was excluded from the merged data.

A prospective data base of an antibiotic-first strategy for acute appendicitis is maintained by the senior author (SH) at VANTXHCS since March 2019. Surgeons at HNSB were interviewed to assess their views on NOM, and their views and responses were recorded as previously reported. ⁶ Once merged, data were dichotomised by hospital. Data are expressed as mean ± SD. PRISM statistical analysis software (GraphPad Software, Inc., San Diego, CA) was used for contingency table analyses and Student’s T-test. Mann–Whitney U-tests and T-tests were used for continuous variables, depending on normalcy. Categorical variables were assessed by Fisher’s exact test or χ ² test. All statistical tests were two-sided, and the statistical significance level was set at a p-value of ≤0.05.

Results

A total of 405 patients were included in this study (148 at HNSB and 257 at VA). In general, VA patients were a substantially higher risk group compared to those from HNSB (Table 1). A more insidious clinical presentation was noted at VA compared to HNSB. Duration of symptoms and heart rate were similar between cohorts. While in both groups the physical examination demonstrated right lower quadrant tenderness at the same rate, physical signs for acute appendicitis were found more frequently at HNSB compared to VA. White blood cell count was higher at HNSB compared to VA. Most (92%) patients at VA underwent CT scanning prior to the operative intervention; only 4% of patients at HNSB underwent ultrasonography to substantiate the diagnosis of acute appendicitis (Table 2). Owing to institutional training, available resources and convention, all appendicectomies performed at HNSB were performed by an open approach, mostly using an incision two-thirds of the distance between the umbilicus and the anterior superior iliac spine (the right iliac fossa). All but 13 interventions at VA were begun laparoscopically, and 12 converted by midline laparotomy to open procedures. The remaining 12 were carried out by a right iliac fossa incision. All patients at HNSB underwent an open approach. The majority (n = 144 (97.3%)) had an initial incision two-thirds of the distance between the umbilicus and the anterior superior iliac spine with seven necessitating and extended incision. In four cases, a median infra-umbilical incision was undertaken.

OPEN IN VIEWER

Table 1.

Patient demographics and co-morbidities at a county hospital in Guatemala and a VA hospital in the United States.

Demographics	HNSB (n = 148)	VANTXHCS (257)	p-Value
Age (years-old) ± SD	32.0 ± 13.65	45.4 ± 15.63	<0.001
Gender (male n (%))	86 (58.5)	233 (90.7)	<0.001
Co-morbidities (n (%))
Dyslipidaemia	0 (0.0)	5 (2.0)	0.09
Diabetes mellitus	3 (3.0)	29 (11.3)	<0.001
Hypertension	2 (1.4)	91 (35.4)	<0.001
Cardiac disease	0 (0.0)	33 (12.8)	<0.001
Renal disease	0 (0.0)	28 (10.9)	<0.001
ASA (n (%))
I–II	138 (98.6)	178 (69.3)	<0.001
III	2 (1.4)	72 (28.0)	<0.001

OPEN IN VIEWER

Table 2.

Clinical presentation, physical exam and diagnostic data at a county hospital in Guatemala and a VA hospital in the United States.

Clinical presentation	HNSB (n = 148)	VANTXHCS (257)	p-Value
Anorexia (n (%))	34 (23.1)	182 (71.7)	<0.001
Nausea and vomiting (n (%))	69 (46.9)	178 (69.5)	<0.001
RLQ pain (n (%))	147 (100)	182 (71.7)	<0.001
Duration of symptoms (h ± SD)	34.5 (31.4)	33.3 (23.2)	0.065
Systolic BP (mmHg ± SD)	119.2 (12.7)	136.4 (17.5)	<0.001
Heart rate (bpm ± SD)	86.0 (13.2)	85.6 (16.9)	0.82
Physical exam (n (%))
RLQ tenderness	144 (97.3)	245 (95.7)	0.42
Rovsing sign	60 (40.5)	58 (22.7)	<0.001
Psoas sign	65 (43.9)	34 (13.3)	<0.001
Obturator sign	26 (17.6)	22 (8.6)	0.007
Diagnostic data
Haemoglobin g/dL ± (SD)	14.4 (6.6)	14.8 (2.3)	0.369
WBC (K/µL) ± SD	14.5 (4.5)	13.3 (4.2)	0.004
Imaging (n (%))	6 (4.1)	236 (92.2)	<0.001

While regional anaesthesia (spinal) was the most common approach in Guatemala (88%), all at the VA underwent general anaesthesia. Operative room time was on average 29 min shorter at HNSB than VA (Table 3). In general, outcomes were superior in patients undergoing an appendicectomy at HNSB compared to VA (Table 3).

OPEN IN VIEWER

Table 3.

Operative approach and outcomes at a county hospital in Guatemala and a VA hospital in the United States.

Operative	HNSB (n = 148)	VANTXHCS (257)	p-Value
Laparoscopic approach (n (%))	0 (0.0)	233 (90.7)	<0.001
Regional (spinal anaesthesia) (n (%))	129 (88.4)	0 (0.0)	<0.001
OR time (min ± (SD))	31.0 (20.1)	60.4 (27.4)	<0.001
Pathology
Perforated appendix (n (%))	70 (47.3)	53 (20.6)	<0.001
Pathology report available (n (%))	1 (0.7)	257 (100)	<0.001
Outcomes
Morbidity (n (%))	2 (1.4)	20 (7.8)	<0.005
Mortality (n (%))	0 (0.0)	1 (0.7)	0.45
30 days re-admission (n (%))	2 (1.4)	23 (9.1)	<0.001
H LOS (days ± (SD))	1.9 (1.3)	2.2 (3.7)	0.31

No patient with a diagnosis of acute appendicitis was ever treated at HNSB by an antibiotic-first strategy, though six at VA were thus treated with two failures. None of the surgeons interviewed at HNSB considered the NOM approach appropriates. ⁶ Nearly, half of their patients were reported to have a perforated appendix.

Discussion

Our study demonstrates that there is a different approach to the surgical management of the same disease in the two contrasting locations well and poorly resourced. The VA patient population is composed of older men with a substantially higher rate of co-morbid conditions and obesity (BMI = 30.3 ± 6.3 kg/m²). Therefore, a laparoscopic approach, being less invasive, has fewer complications. ⁸

On the other hand, the patient population at HNSB is younger and healthier and had a little less than half (42%) female, all of which factors are associated with a reduced risk of complications. Complications from the standard open approach at HNSB are substantially lower (1.4%) with zero mortality. Thus, laparoscopy, on this basis alone would not justify a change in strategy; the training as well as the cost of technology that would accompany the introduction of laparoscopy also mitigates against its use. Further, the operative cost would be higher as open appendicectomy at HNSB took half as long as laparoscopically at VA (31.0 vs. 60.4 min). More importantly, a major barrier to laparoscopy at HNSB is the need for general anaesthesia, where limited supplies of anaesthetic agents, reserved for more major abdominal operations, exist.

Whilst NOM is a strategy that has generated substantial controversy in USA, several randomised controlled trials and systematic reviews have documented its feasibility.^3,9–14 At HNSB, patients tend to present with more advanced disease compared to those at VA, with more frequent positive clinical signs and higher leukocytosis at presentation. Intravenous antibiotics are in limited supply at HNSB, and once discharged, patients need to seek medication at a pharmacy; thus, compliance is uncertain. Additionally, extra hospital days required by NOM would take up resources of an already overly occupied hospital. Surgeons also noted that NOM is not readily accepted by their patient population, who expect surgical intervention for acute appendicitis, and the bear liability if complications should ensue with NOM.

Currently, at HNSB, almost all (90%) elective cholecystectomy is performed open. Barriers preventing incorporation of laparoscopy to HNSB have been outlined.^5,15 Whilst there is a great deal of enthusiasm by all participating surgeons in implementing laparoscopy for gallbladder disease, our data show that for appendicitis, laparoscopy and an antibiotic-first therapy are not consistent with best system-based practices at HNSB and are currently not welcome at HNSB.

Our study has substantial limitations. It is purely observational and is severely limited by extracting data from paper charts at HNSB and, given the retrospective nature of the study, is prone to selection bias and collection errors. Further, not all charts at HNSB had complete information. Thus, it is unclear how certain outcomes with incomplete data would change the statistical results. Also, the number of patients in this cohort is relatively small, and lack of statistical significance could be the result of a type II error. However, this study shows that the patient population, available resources and surgical training dictate different system-based practices to an appendicectomy with acceptable outcomes in both low-income and high-income countries.