Introduction
Acute appendicitis is a common surgical condition and one that impacts the global burden of surgical disease.(1) It remains one of the most common indications for non-elective surgery worldwide. The timing of treatment of appendicitis has been a contentious issue for the past 120 years. Historically, surgeons adopted an aggressive policy of early surgical intervention without delay when the diagnosis was suspected.(2) This view emanated from the conventional pathophysiologic model of acute appendicitis, which was based on a relationship between time and disease progression. Thus, any delay in surgery was believed to result in an acute, uncomplicated, inflamed appendix progressing to necrosis and perforation.(3) While it is universally accepted that if left untreated, acute appendicitis may have dire consequences, the resort to emergency surgery for appendicitis has in recent years been challenged by evidence from numerous studies, which have focused on defining a safe time frame by which surgery for acute appendicitis can be delayed without any increase in morbidity.(4,5) On the other end of the spectrum, numerous studies still maintain that any delay in surgery should be avoided.(6) This study aimed to determine if surgery for patients presenting with acute unperforated appendicitis clinically can be safely delayed for a short period without resulting in adverse outcomes.
Methods
This was a retrospective descriptive cohort study based on a record review conducted at Sebokeng Regional Hospital, a public hospital in South Africa.
All patients 18 years and older identified from the Emergency Surgery record database as having undergone a laparoscopic or open appendicectomy for uncomplicated acute appendicitis were included. Patients having undergone an exploratory laparotomy for suspected perforated appendicitis and missing patient files were excluded from the outset.
The diagnosis of acute appendicitis was initially made only using the Alvarado score and then confirmed at surgery. The Alvarado score is a widely known, cost-efficient method combining clinical and laboratory findings to predict the likelihood of acute appendicitis.(7) All patients included in this study had an Alvarado score greater than 7.
Unperforated acute appendicitis was clinically defined as patients presenting with localized right iliac fossa (RIF) pain, localized tenderness, or rebound tenderness on clinical examination without generalized peritonitis or a right iliac fossa mass (suspicious for an appendiceal mass).
Symptom onset was defined as when the patient first noted the most discerning feature of acute appendicitis, abdominal pain. Pre-hospital delay was defined as the patient's duration of symptoms from onset until presentation at Sebokeng Regional Hospital. The in-hospital delay was divided into three categories: time from presentation to the hospital until assessment by the emergency department (ED) casualty officer, time taken to be assessed by the surgeon on call, and time from the decision to operate until surgery.
No patient was considered for upfront non-operative management. The proven uncomplicated status of acute appendicitis is a prerequisite for any conservative approach. While guidelines do not clearly state how one may do this, preoperative imaging appears to be an essential tool in differentiating uncomplicated from complicated appendicitis.(8) This was impossible in a resource-constrained environment such as Sebokeng Regional Hospital, where access to imaging modalities such as abdominal ultrasonography and CT scan to confirm the diagnosis was limited.
The severity of intraoperative appendicitis was classified according to the American Association for the Surgery of Trauma (AAST) grading system for acute appendicitis. For this study, patients were classified as having either mild appendicitis (AAST I), gangrenous appendicitis (AAST II), or locally perforated appendicitis (AAST III+IV).
Patient files were used to track the patient's post-operative course, assess the development of complications as defined by the Clavien-Dindo classification system, determine in-hospital mortality, and determine the total length of stay until discharge.
Statistical analysis
Stata version 13.1 was used for statistical analysis. Descriptive statistics were used to describe the patient population, pre- and in-hospital delay, and surgical outcomes, namely, severity of appendicitis, morbidity, and length of hospital stay. Univariable and multivariable Poisson regression with robust error variance was used to determine factors associated with suppurative or gangrenous appendicitis. A p-value <0.05 was considered statistically significant. Factors with a p-value <0.2 in the univariable analyses were included in the multivariable model.
Ethical clearance for this study was obtained from the University of the Witwatersrand's Human Research Ethics Committee.
Results
A total of 103 patients were eligible for inclusion in this study. The median age was 25 (IQR 21-33); 60 (58%) were male, and 92 (89%) were of African descent. All 103 (57%) patients had an Alvarado score >7. The median white cell count (WCC) was 13 cells/ul (IQR 10-17 cells/ul).
Pre- and in-hospital delay and surgical outcome
The median pre-hospital delay was 48 hours (IQR 24-72 hours), time from presentation to the hospital until assessment by the emergency department casualty officer 1.33 hours (IQR 0.25–2.5 hours), time to be assessed by the surgeon on call 2.33 hours (IQR 1.50–4.17 hours) and time from the decision to operate to surgery 10 hours (IQR 5.50–16.33 hours) (Table 3.1).
Analysis of overall delay before surgery (N = 103)
| Time period | Median (IQR) |
|---|---|
| Pre hospital delay (in hours) | |
| Onset of symptoms to presentation | 48 (24 – 72) |
| In hospital delay (in hours) | |
| Presentation to casualty officer | 1.33 (0.25 – 2.5) |
| Casualty officer to surgeon | 2.33 (1.50 – 4.17) |
| Surgeon to surgery | 10 (5.50 – 16.33) |
| Overall hospital delay | 15.75 (9.5 – 22.42) |
| Total delay (in days) | 2.46 (1.70 – 3.58) |
A pathologist evaluated all appendix specimens. Thirty-six (35%) patients had mild appendicitis, 25 (24%) patients had gangrenous appendicitis, and 42 (41%) patients had suppurative appendicitis. There were no statistically significant differences in the overall delay before surgery with the histological findings. However, differences existed in the pre-hospital delay when comparing the mild appendicitis group versus the suppurative/gangrenous appendicitis group. The pre-hospital delay was longer among the suppurative/gangrenous appendicitis group compared with the mild appendicitis group (Median 48 hours versus 24 hours) (p = 0.032) (Table 3.2). There were no deaths in this cohort of patients. Post-operative complications were classified according to the Clavien-Dindo Classification system (Figure 3.1). 35 (34%) patients developed complications, all of them arising from patients found to have suppurative/ gangrenous appendicitis. Of the 67 (65%) patients found to have suppurative/gangrenous appendicitis, 32 (47%) patients did not develop any complications. Of the 35 (53%) patients that did develop complications, 20 (29.9%) patients developed surgical site infection with intra-abdominal collections requiring a relook laparotomy, and 11 (16.4%) patients developed a post-operative ileus (Table 3.3). A pre-hospital delay of 2 days was associated with postoperative complications (p = 0.002). Patients who developed postoperative complications were taken to the theatre much sooner than those who did not (p = 0.030) (Table 3.4). The median (IQR) length of hospital stay for patients with mild appendicitis was 1(1-2) days, and for patients with suppurative/gangrenous appendicitis, 4 (3-6) days (p < 0.001).
Clavien-Dindo classification of post-operative complications for mild, suppurative and gangrenous appendicitis.
Analysis of the association between overall delay before surgery and the histology/ pathology of the appendix (N = 103)
| Time period | Early mild AppendicitisMedian (IQR) | Gangrenous/suppurative Median (IQR) | P value (Wilcoxon Ranksum test) |
|---|---|---|---|
| Duration of symptoms (in hours) | 24 (24–48) | 48 (24–72) | 0.032 |
| Presentation to casualty officer (hours) | 1.5 (0.54–2.08) | 1.25 (0–2.5) | 0.770 |
| Casualty officer to surgeon (hours) | 2 (1.12–4.50) | 2.5 (1.5–4.0) | 0.675 |
| Surgeon to surgery (hours) | 13.25 (6.5–19.17) | 9.42 (5.33–15.17) | 0.089 |
| Overall hospital delay (hours) | 19.33 (10.04–23.83) | 13.83 (8.83–20.17) | 0.145 |
| Total delay (in days) | 1.99 (1.76–3.01) | 2.58 (1.67–3.91) | 0.120 |
Type of complications in patients with suppurative/ gangrenous appendicitis.
| Type of complication in patients with suppurative/ gangrenous appendicitis | N | % |
|---|---|---|
| Grade 1-Ileus | 11 | 16.4 |
| Grade 1-Wound sepsis managed with dressings | 1 | 1.5 |
| Grade 2-Aspiration pneumonia | 1 | 1.5 |
| Grade 3b-Iatrogenic bowel injury | 1 | 1.5 |
| Grade 3b-Surgical site infection- relook and washout for collections | 20 | 29.9 |
| Grade 4a-PE, Biventricular failure | 1 | 1.5 |
| None | 32 | 47.8 |
Association of overall delay before surgery and morbidity.
| Time period | No post-op morbidity | Post-op morbidity | P value |
|---|---|---|---|
| Onset of symptoms to presentation (hours) | 24 (24–48) | 48 (24–96) | 0.002 |
| Presentation to casualty officer (hours) | 1.08 (0.21–2.08) | 2 (0.25–2.75) | 0.244 |
| Casualty officer to surgeon (hours) | 2.33 (1.5–4.08) | 2.33 (1.5–4.33) | 0.884 |
| Surgeon to surgery (hours) | 11.54 (6.5–17.96) | 7.33 (4.83–15.17) | 0.030 |
| Overall hospital delay (hours) | 16.62 (10.75–23.62) | 13.83 (8.0–19.17) | 0.123 |
| Total delay (in days) | 2.09 (1.62–3.06) | 2.80 (2.14–4.35) | 0.056 |
Factors associated with suppurative or gangrenous appendicitis
Univariable and multivariable Poisson regression analysis showed that the overall delay before surgery was not associated with suppurative or gangrenous appendicitis. However, age and white cell count were identified as independent predictors of suppurative or gangrenous appendicitis (Table 3.5). A WCC ≥11cells/ul was associated with a 55% higher risk of suppurative/ gangrenous appendicitis (adjusted relative risk (aRR) 1.55 [95% confidence interval (CI) [1.09,2.20] (p = 0.016) compared to a low or normal white cell count. In comparison, there was a trend toward a higher risk of suppurative or gangrenous appendicitis with age older than 25 years (adjusted relative risk (aRR) 1.34 [95% confidence interval (CI) [0.99,1.80] (p = 0.052). Higher pre-operative temperatures and CRP levels were not associated with an increased risk of suppurative/gangrenous appendicitis.
Factors associated with suppurative/ gangrenous appendicitis.
| Variable | % with suppurative/ gangrenous appendicitis | Univariable Relative risk (RR) 95% CI | p-value | Multivariable Relative risk (RR) 95% CI | p-value |
|---|---|---|---|---|---|
| Total delay (hours) | |||||
| <48 | 24/44 (54.6) | 1.00 | 1.00 | ||
| 48–72 | 18/25 (72.0) | 1.32 (0.92–1.90) | 0.137 | 1.32 (0.92–1.90) | 0.129 |
| 72–96 | 8/14 (57.1) | 1.05 (0.62–1.78) | 0.864 | 1.02 (0.64–1.62) | 0.931 |
| > = 96 | 17/20 (85.0) | 1.56 (1.12–2.16) | 0.084 | 1.28 (0.92–1.77) | 0.139 |
| White cell count (cells/µL) | |||||
| <11 cells/µl | 19/42 (45.2) | 1.00 | 1.00 | ||
| ≥11 cells/µl | 48/61 (78.7) | 1.74 (1.21–2.49) | 0.003 | 1.55 (1.09–2.20) | 0.016 |
| C-reactive protein (mgl/L) | |||||
| <10 | 9/25 (36.0) | 1.00 | 1.00 | ||
| 10–50 | 12/21 (57.1) | 1.59 (0.83–3.02) | 0.159 | 1.26 (0.67–2.37) | 0.471 |
| ≥50 | 46/57 (80.7) | 2.24 (1.31–3.85) | 0.003 | 1.61 (0.90–2.86) | 0.109 |
| Temperature (°C) | |||||
| <37 | 62/97 (63.9) | 1.00 | 1.00 | ||
| ≥37 | 5/6 (83.3) | 1.30 (0.88–1.93) | 0.182 | 1.00 (0.69–1.47) | 0.989 |
| Age (years) | |||||
| <25 | 26/49 (53.1) | 1.00 | 1.00 | ||
| ≥25 | 41/54 (75.9) | 1.43 (1.06–1.94) | 0.021 | 1.34 (0.99–1.80) | 0.054 |
Discussion
Acute appendicitis affects both genders, does not discriminate between races, and affects both young and old. Its incidence peaks in children and young adults.(9) Our study depicts this, as 71% of patients were in the 18-30 age range. Most patients were of African descent. A study conducted at another regional hospital in South Africa reflected similar demographic findings.(10)
The main known pathogenic trigger of acute appendicitis is obstruction of the appendicular lumen, which leads to inflammation. Stasis in the lumen allows intestinal bacteria to multiply, recruiting white blood cells and accumulating purulent material. Obstructing the lumen of the appendix may result from several causes, including faecoliths, lymphoid hyperplasia, foreign bodies, parasites, and primary and metastatic tumours. Obstruction causes the intraluminal pressure to rise, resulting in wall ischaemia, necrosis, and eventual rupture.(9) If this process is left untreated, progressive inflammation from this seemingly benign appendage will result in disastrous consequences. This is directly related to the development of complicated intra-abdominal infections such as intra-abdominal abscesses or purulent peritonitis, a process that is time-dependent and requires prompt source control to avoid significant morbidity and mortality.(11)
Although performing an appendicectomy is considered the bread and butter of any surgeon's armamentarium, its effects and the effects of delays have remained contentious over the years. This study demonstrated that while overall delay before surgery, encompassing both pre- and in-hospital delay, did not affect the outcome of acute appendicitis, pre-hospital delay was a significant factor in both disease severity and patient outcome.
Pre-hospital delay
Contacting a health care system is referred to as health-seeking behaviour. It is defined as “any action or inaction undertaken by individuals who perceive themselves to have a health problem or to be ill to find an appropriate remedy.”(12) The median pre-hospital delay in this cohort of patients was two days. This study demonstrated that pre-hospital delays of more than two days were more likely to have suppurative or gangrenous appendicitis on histology and thus were more likely to develop post-operative complications with a resultant longer length of hospital stay. A similar study conducted in Malawi showed that while delays in accessing health care did not directly increase the risk of mortality, it was an important factor driving non-fatal complications.(13) The majority of patients with suppurative or gangrenous appendicitis who developed complications in our study had surgical site infections with underlying intra-abdominal collections. Source control is thus essential in these patients. In our cohort of patients, surgical intervention in the form of a relook laparotomy was undertaken to achieve source control. An important conclusion from our findings is that patients with a pre-hospital delay of more than two days should be fast-tracked for surgery.
Delays in seeking access to medical care are common to our patient population. A study at a peri-urban public hospital in Kwa-Zulu Natal found that 70% of patients presenting with acute appendicitis were delayed more than 48 hours.(14) Another local study by Yang et al. compared patients with acute appendicitis in the private and public health sectors in Cape Town, South Africa, and found that pre-hospital delays were comparable at 49 and 56 hours, respectively.(15)
Several factors could be implicated in delays in seeking medical care. As a low- and middle-income country (LMIC), South Africa has a social environment and population demography that is uniquely different from high-income countries. In addition, disparities exist between urban and rural patients seeking medical care, and differences exist between private and public sector hospitals within our country. The former was highlighted by a study conducted in Pietermaritzburg that identified rural origin as an independent predictor of worse outcomes in patients with acute appendicitis.(16) Significant barriers to health care could be extrapolated from other countries with similar social standing.(17,18)
In hospital delay
Reasons for in-hospital delays before surgery are multifactorial. Organizational factors such as the efficiency and staffing of the emergency room, the use of an effective triage system, the number of surgical doctors on call, their workload, availability of radiological services, and access to a 24-hour emergency operating theatre, amongst others, play a pivotal role. A study in Sudan found that the misdiagnosis of acute appendicitis in 65% of the study sample contributed to the most significant reason for delayed surgery.(19)
In our cohort of patients, the in-hospital delay was divided into three categories: time from presentation to hospital until assessment by the casualty officer, the time taken to be assessed by the surgeon on call, and the time from the decision to operate to surgery. While none of these categories of in-hospital delays reached statistical significance, a few observations could be drawn. The median delay for each category was 1 hour, 2 hours, and 10 hours, respectively. Longer delays to surgery after being diagnosed with acute appendicitis could be explained by the fact that there was only one senior surgeon on call every night, assisted by a junior staff comprising interns or community service medical officers. This team was responsible for dealing with all new surgical emergencies and tending to acutely deteriorated surgical inpatients. Thus, the surgeon's prerogative was on call to triage all cases awaiting theatre. One could assume that critically ill patients, after a period of adequate resuscitation, were prioritized above more stable patients. However, what did reach statistical significance was that patients retrospectively found to have suppurative or gangrenous appendicitis were taken to the theatre much sooner. Thus, one could postulate that the surgeon acted quicker for sicker patients.
This retrospective study did not find that the overall in-hospital delay affected the outcome of acute appendicitis. In a systematic review and meta-analysis conducted by Bhangu et al., the main finding was that a short delay of 12-24 hours was not associated with an increased risk of complicated appendicitis.(5) In addition, the meta-analysis showed that these short delays were not associated with increased rates of infectious complications.(5) The study showed that delays beyond 48 hours were associated with an increased risk of surgical site infection without a risk of perforation.(5) The investigators suggested short in-hospital delays are a safe option, especially with intravenous antibiotics and fluid resuscitation. However, prompt surgery (if resources allow) should still be undertaken to allow for faster resolution of patient symptoms.(5) Another meta-analysis demonstrated that a delay of up to 24 hours post-admission for suspected uncomplicated acute appendicitis was not a risk factor for complicated appendicitis, postoperative surgical site infection, or morbidity.(4)
While emergency surgery is a life-saving necessity, it is not without drawbacks. Safely delaying appendicectomies for daylight hours could have benefitted our setting considerably. Our healthcare system has numerous challenges ranging from lack of equipment, staff shortages, and, at times, insufficient senior supervision. Daytime operating would enhance the training of junior surgeons and improve patient outcomes compared to night-time emergency operations performed by trainees without supervision. A safe time frame for delaying appendicectomies in our setting could not be determined, and a possible reason for this could be that this was a single-centre retrospective study with a small sample size. Thus, more hospitals with a larger pool of patients could be a better representation when answering this objective.
Our study found that age and WCC were independent predictors of gangrenous and suppurative appendicitis. A survey by Nomura et al. also identified age as a risk factor since aging is associated with compromised immunity and more severe inflammation.(20) However, since a minority of our patient cohort was above 41, extremes of age could not be implicated in this process.
No deaths were reported in our cohort of patients. A possible reason for this could be that patients with complicated appendicitis or those presenting with severe sepsis were excluded from our study from the outset.
Study limitations
This study has several limitations. First, it was a retrospective study, so limited control could be exercised over meticulous record-keeping. Second, the generalizability of the results is limited as our study was conducted at a single regional hospital. Therefore, it is uncertain whether such findings could be extrapolated to tertiary-level teaching hospitals with a more extensive senior staff complement and with more supervision. Lastly, many patients were excluded from this study due to missing records, significantly impacting the sample size.
Conclusion
This study could not determine a safe time frame for delaying appendicectomies. Public health and sociological research encompassing a more significant number of hospitals with a larger sample size would provide information to improve our understanding of the reasons for late presentation in our setting. However, our study did demonstrate the impact of pre-hospital delay on surgical findings and subsequent outcomes of acute appendicitis in a resource-constrained setting, as we found a significant correlation between time to presentation and increased morbidity. Based on the study findings, patients older than 25 years with a pre-hospital delay of >2 days and an elevated white cell count > 11 cells/µl on presentation should be fast-tracked for surgery.