Non reducible inguinal hernias in Malawi: an occupational hazard

Abstract

Purpose:

Inguinal hernias are one of the most common surgical conditions worldwide. Due to limited surgical access in low- and middle-income countries, many hernias present emergently; however, data on the resultant outcome disparities is limited. We, therefore, sought to describe the epidemiology, clinical features, and outcomes of incarcerated inguinal hernias at a tertiary center in Malawi.

Methods:

This is a retrospective analysis of the acute care surgery registry at Kamuzu Central Hospital in Lilongwe, Malawi. All patients > 13 years admitted with a non-reducible inguinal hernia from 2013–2019 were included. The primary outcome was in-hospital mortality. A Poisson multivariable regression determined factors associated with increased risk of mortality.

Results:

A total of 297 patients presented with non-reducible inguinal hernias, the majority of which were young (median age 38), male (93.6%), farmers (47.8%). Of the 81% who underwent surgery, 55% were delayed ≥ 24 hours. 19.5% of hernias were strangulated. Overall mortality was 5.4%. Increased age (RR 1.06, 95% CI 1.01–1.12), shock index ≥ 1 (RR 4.82, 95% CI 1.45–16.09), and delay ≥ 24 hours from presentation to operative intervention (RR 11.24, 95% CI 1.55–81.34) resulted in an increase in relative risk of mortality.

Conclusion:

Non-reducible inguinal hernias largely affect young male farmers in Malawi. Delays to care can limit economic productivity for this rural population, as well as, yield considerable risk of mortality. While specific patient and institutional factors must be further elucidated, increased awareness, public health prioritization, and surgical capacity building is needed to reduce further hernia-related morbidity and mortality.

Introduction

Inguinal hernias are a common surgical condition with more than 20 million people worldwide undergoing an inguinal herniorrhaphy each year [1]. The estimated lifetime risk of developing an inguinal hernia is approximately 27–43% in men and 3–6% in women with the global prevalence of inguinal hernias estimated to be 2–9%; however, data on inguinal hernias are largely derived from population-based studies in high-income countries (HIC) [2, 3]. As disparities in surgical access and capacity result in a large proportion of untreated inguinal hernias in low- and middle-income countries (LMIC), the prevalence of inguinal hernias in sub-Saharan Africa has been estimated to be much higher than these approximations, ranging from 8–20% [4].

Within rural populations in Africa, untreated inguinal hernias result in a high burden of morbidity, mortality, and poverty. While the exact incidence of inguinal hernia repair in Africa is unknown, prior studies have estimated that 175 inguinal hernia repairs are needed per 100,000 population per year, but only 25 per 100,000 are estimated to currently be performed [5]. As a result of this unmet need, there is an increased risk of non-reducible, or incarcerated hernias and strangulation, and thus high rates of emergent hernia surgery. As only 4 of the estimated 30 strangulated hernias per 100,000 population in Africa that require emergent repair are performed, these surgical disparities also result in high rates of mortality [6].

In response to the World Health Organization’s call for a global prioritization of surgical disease, inguinal hernias have been proposed as a “priority 1 surgical condition” due to its high burden of disease and the feasibility of repair [7]. Given the impact of untreated inguinal hernias particularly on an agrarian society, describing this unmet need is imperative to public health prioritization. To address this knowledge gap, we sought to characterize the epidemiology and outcomes of patients presenting with non-reducible inguinal hernias in Malawi.

Methods

This is a retrospective analysis of prospectively collected data in the Kamuzu Central Hospital (KCH) acute care surgery (ACS) registry in Lilongwe, Malawi. KCH is a 900-bed tertiary hospital that serves 8 districts in the central region of Malawi with a catchment population of approximately 6 million people. All patients over 13 years of age who presented to KCH from September 2013 to August 2019 with the admission diagnosis of incarcerated inguinal hernia were eligible for inclusion. No patients were excluded. Patients were classified as non-reducible if the inguinal hernia content could not be pushed back into the peritoneal cavity by the clinical provider and documented as such. In our study, all surgical patients were managed with open herniorrhaphy, of which the Bassini repair was the method of choice in the majority of operations. Mesh hernioplasty is not performed at our center due to the cost, lack of availability, and lack of training [8].

Descriptive statistics were used to characterize the demographic and clinical features of patients presenting with non-reducible inguinal hernias to KCH. The mean and standard deviation (SD), median and interquartile range (IQR), and shape of the distribution were calculated for each continuous variable and frequencies were tabulated for categorical variables. Baseline demographics included age, sex, occupation, and known co-morbidities. Clinical characteristics included transfer status, type of surgery performed, time to surgery, strangulation, post-operative complications, vital signs, and the presence of shock. Time to surgery was defined as the time from presentation at any healthcare facility to surgery at KCH and was dichotomized between less than 1 day and ≥ 1 day. Strangulation was defined as an intraoperative finding of vascular compromise to the hernia contents. Lastly, a dichotomous variable for shock was created using shock index, the ratio of heart rate/systolic blood pressure, which has been shown to be a useful predictor of poor hemodynamic status when ≥ 1 in septic patients [9]. Unfortunately, approximately 30% of vital sign data was missing, which is not uncommon in the LMIC setting. Sensitivity analysis showed no significant difference in the proportion of shock index missing by mortality (p=.957), thus these values were likely missing at random.

The primary outcome of interest was in-hospital mortality. To determine the factors associated with mortality, bivariate analysis of all patients who presented with non-reducible inguinal hernias was conducted. Factors associated with mortality were compared between survivors and non-survivors using student’s t-test, Kruskal Wallis test, and Pearson’s X² test when appropriate. A p-value < 0.05 was considered a statistically significant difference between survivors and non-survivors. A modified univariable and multivariable Poisson regression was performed to determine the factors associated with mortality in this population [10]. Variables with a p-value < 0.05 in the bivariate analysis and those selected a priori were included in the univariable regression. Statistically significant and a priori variables from the univariable regression were then included in the multivariable analysis.

Results

During this 6-year study period, 297 patients presented to KCH with non-reducible inguinal hernias, representing 7.7% of the entire ACS population (Table 1). The median age was 38 years (IQR 28–52) and the vast majority were male (n = 278 [93.6%]). Non-reducible inguinal hernia patients were most frequently farmers (n = 141[47.8%]), followed by unemployed (n = 39 [13.2%]) and small business owners (n = 34 [11.5%]). Most did not have any known comorbidities (n = 191 [90.5%]), but some had HIV (n = 11 [3.7%]) and malaria (n = 11 [3.7%]).

Table 1.

Characteristics of patients presenting to Kamuzu Central Hospital with an incarcerated inguinal hernia.

	Incarcerated Inguinal Hernias (n = 297)
Age (years), median (IQR)	38 (28–52)
Male, n (%)	278 (93.6)
Occupation, n (%)
Peasant Farmer	141 (47.8)
Small Business Owner	34 (11.5)
Student	22 (7.5)
Homemaker	14 (4.8)
Construction	10 (3.4)
Office/Bank worker	9 (3.1)
Unemployed	39 (13.2)
Comorbidities, n (%)
Hypertension	3 (1.0)
Malaria	11 (3.7)
None	191 (64.3)
HIV	11 (3.7)
Missing/unknown	81 (27.3)
Initial HR, n (%)
≤ 100	170 (57.2)
> 100	37 (12.5)
Missing	90 (30.3)
Initial SBP, n %
≤ 90	4 (1.4)
> 90	202 (68.0)
Missing	91 (30.6)
Initial RR, n (%)
< 20	12 (4.0)
≥ 20	109 (36.7)
Missing	176 (59.3)
Shock index, n (%)
Shock (index ≥ 1)	11 (3.7)
Not in shock (index < 1)	195 (65.7)
Missing	91 (30.6)
Transferred, n (%)	213 (72.0)
Admission disposition, n (%)
Discharged	1 (0.3)
Ward	296 (99.7)
ICU	0 (0)
Underwent an imaging study on arrival a, n (%)	94 (31.7)
Underwent surgery, n (%)	243 (81.8)
Disposition of those that did not undergo surgery, n (%) [N=53]
Discharge	42 (79.2)
Abscond	9 (17.0)
Death	2 (3.8)
Time from presentation at any health care facility to operative intervention, n (%)
< 1 day	1,084 (44.9)
≥ 1 day	1,332 (55.1)
Strangulated inguinal hernia, n (%)	58 (19.5)
Bowel perforation, n (%)	2 (0.7)
Ileostomy or colostomy	3 (1.0)
Post-op Complications, n (%)
Take back	3 (1.0)
Wound infection	3 (1.0)
Disposition, n (%)
Death	16 (5.4)
Discharge	269 (90.9)
Abscond	10 (3.4)
Transfer	1 (0.3)

^aX-Ray or ultrasound

The majority of non-reducible inguinal hernia patients at KCH were hemodynamically stable on presentation with only 12.5% presenting with tachycardia (heart rate > 100 bpm) and only 1.4% hypotensive (systolic blood pressure < 90 mmHg) (Table 1). Based on shock index ≥ 1, only 3.7% of non-reducible hernia patients presented in shock. Most were transferred from outside facilities (n=213 [72.0%]) and the vast majority were admitted to routine floor care beds (n = 296 [99.7%]). One-third of patients in this cohort underwent a pre-operative imaging study (n = 94 [31.7%]), consisting of either an abdominal radiograph or ultrasound.

Eighty-one percent of patients with non-reducible inguinal hernia patients were managed operatively. Of those that did not undergo surgery (n = 54 [18.9%]), the majority were discharged following hernia reduction (n = 42 [79.3%]) with the remainder absconding (n = 9 [17.0%]) or dead (n = 2 [3.8%]). For those that underwent surgery, time from presentation to a health care facility to operative intervention exceeded 24 hours in 55.1% of this cohort. While the majority underwent uncomplicated inguinal hernia repair, some were found to have evidence of strangulation intra-operatively (n =58 [19.5%]), with bowel perforation occurring in 2 (0.7%) patients. In the case of strangulation, bowel resection was predominantly performed with primary anastomosis; however, 3 (1.0%) patients underwent ileostomy or colostomy creation. Post-operative complications were rare. Only 3 patients (1.0%) required a take back to the operating room and 3 (1.0%) developed a documented wound infection. Mortality in this population was 5.4%.

Bivariate analysis of clinical and demographic factors associated with mortality are shown in Table 2. Statistically significant differences between survivors and non-survivors included age, the presence of shock, and time to surgery. Univariable analysis of these factors indicated that increasing age (RR 1.05, 95% CI 1.02–1.08, p = 0.001) and prolonged time to surgery (RR 4.87, 95% CI 1.39–17.04, p = 0.01) were associated with an increased relative risk of mortality (Table 3). While shock (RR 3.94, 95% CI 0.96–16.14, p = 0.06) and transfer status (RR 1.21, 95% CI 0.99–1.48, p= 0.07 also showed greater risk of mortality, these factors were not statistically significant. Interestingly, strangulation did not affect mortality in this population (RR 0.95, 95% CI 0.28–3.23, p = 0.9). In the multivariable analysis, increased age (RR 1.06 [95% CI 1.01–1.12], p = 0.02), the presence of shock (RR 4.82 [95% CI 1.45–16.09], p = 0.01), and prolonged time to operative intervention resulted in a statistically significant increase in relative risk for mortality, with surgical delay ≥ 1 day having the highest associated risk (RR 11.24, 95% CI 1.55–81.34, p = 0.02) (Table 3).

Table 2.

Bivariate analysis of all incarcerated inguinal hernias by mortality.

	Survived n = 281 [94.6%]	Died n = 16 [5.4%]	p-value
Age (years), median [IQR]	36 (28–50)	58 (41–73)	0.002
Female sex, n (%)	16 (5.7)	3 (18.8)	0.04
HIV positive, n (%)	10 (3.6)	1 (6.3)	0.9
Presented in shock, n (%)	9 (4.6)	3 (18.2)	0.05
Shock index, mean (SD)	0.9 (0.3)	0.7 (0.2)	0.002
Underwent an imaging study, n (%)	89 (31.7)	5 (31.3)	1.0
Transferred to KCH, n (%)	198 (70.7)	15 (93.8)	0.1
Time to surgery from arrival at any health care facility, n (%)			0.006
< 1 day	135 (59.2)	3 (21.4)
≥ days	93 (40.8)	11 (78.6)
Bowel perforation, n (%)	2 (0.7)	0 (0.0)	0.7
Strangulated inguinal hernia, n (%)	55 (19.6)	3 (18.8)	0.9
Take back, n (%)	3 (1.1)	0 (0.0)	0.7
Wound infection, n (%)	3 (1.1)	0 (0.0)	0.7

Table 3.

Modified univariable and multivariable Poisson regression of factors associated with mortality in patients presenting with incarcerated inguinal hernias

	Univariable analysis		Multivariable analysis
	RR (95% CI)	p- value	RR (95% CI)	p-value
Age (years)	1.05 (1.02–1.08)	0.001	1.06 (1.01–1.12)	0.02
Presented in shock	3.94 (0.96–16.14)	0.06	4.82 (1.45–16.09)	0.01
Transferred	1.21 (0.99–1.48)	0.07
Time to surgery from arrival at any health care facility, n (%)
< 1 day	ref
≥ 1 days	4.87 (1.39–17.04)	0.01	11.24 (1.55–81.34)	0.02
Strangulated hernia	0.95 (0.28–3.23)	0.9

Discussion

Inguinal hernia repair is one of the most common surgical operations worldwide accounting for 10 to 15% of all surgical procedures and is the second most common surgical procedure after appendicectomy [11]. However, the prevalence in sub-Saharan Africa is poorly described. In this hospital-based study of patients presenting with non-reducible inguinal hernias, we show that the incidence of non-reducible hernia is higher in young men, who are typically engaged in physical labor, particularly farming, with an overall mortality of 5.4%. Predictors of increased risk of mortality were increasing age, the presence of shock on presentation, and delay in operative intervention over 24 hours from presentation.

The demographic characteristics and clinical presentation of inguinal hernias in our cohort are comparable to other hernia studies in sub-Saharan Africa [12, 13]. In Nigeria, patient characteristics are similar with over 60% of inguinal hernia patients being agricultural workers and the vast majority were males in early adulthood [14]. Furthermore, there is an association between the prevalence of inguinal hernia and low socioeconomic class [15], reflected in our study population of largely subsistence farmers with no formal education that are living in rural areas. Additional studies from sub-Saharan Africa have demonstrated the implication of low socioeconomic status on the hernia population as decreased access to health care facilities, poor awareness of the disease, and cost have been cited as the most common reasons for failure to seek timely medical care [16]. As a result of these delays to care, the complication profile in our population was higher than that of HIC. While 19.5% of our study population was found to have evidence of strangulation with 13% requiring bowel resection, two randomized control trials comparing watchful waiting and elective repair in HIC have shown the risk of strangulation to be only 2.8 – 4.5% for those presenting with non-reducible inguinal hernias [17]. Fortunately, post-operative complications such as wound infection and return to the operating room were rare, which was similar to other populations [2, 18].

Mortality rates vary throughout the region; however, our findings are similar to estimates from Ghana and Nigeria at 6.2% and 5.3%, respectively [5]. In comparison, mortality rates following emergent hernia repair in the United States are estimated at 3.3% [19]. Decreased access to care is multifaceted in LMIC, reflective not only of the physical lack of a surgical ecosystem, operating theaters, and trained surgical and anesthesia personnel; but also, the associated barriers to transportation, the financial impact of lost days of work, and decreased awareness of the disease. This phenomenon has been characterized by the Lancet Commission on Global Surgery as the “three delays to care”, first the patient’s delay in seeking care, followed by the limited access to a nearby health care facility, and finally lack of adequate surgical capacity at the health care facility [20].

Patient factors related to delay in care include cost, lack of education, and reliance on informal practitioners [21]. The accessibility of hospitals is also of significant concern as an average of 29% of the population in Africa live more than 2 hours from a hospital equipped with emergency care [22]. If able to reach a hospital, the limited surgical workforce remains a significant barrier to safe, effective surgical care delivery in LMIC. It is estimated, that there are as little as 0.5 surgeons per 100,000 people sub-Saharan Africa compared to 71 surgeons per 100,000 people in HIC [23, 24]. Addressing both the patient factors related to delay in care and promoting surgical capacity building are necessary to addressing this disparity in both hernia and all surgical care.

This study is limited by the retrospective design. Vital sign data was often missing, preventing calculation of shock index in a number of patients. Finally, in-hospital complications were captured leaving long-term complications like hernia recurrence unavailable to analyze. Despite these limitations, we were able to capture a relatively comprehensive set of patient characteristics in comparison to similar hospital-based studies in LMIC serving to highlight the burden of non-reducible inguinal hernias in Malawi and the necessity of future research on the unmet surgical needs of this population.

Conclusion

Non-reducible inguinal hernias represent a significant proportion of the surgical burden of disease in Malawi. Delays to care yield considerable morbidity and mortality for this rural population of young, male farmers, and can result in loss of economic productivity. While patient and institutional factors specific to this population must be further elucidated, increased patient awareness and improved access to safe, effective surgical care is imperative to reduce further hernia-related impoverishment, morbidity, and mortality.