Early Postoperative Deaths in the Surgical Intensive Care Unit of CHU-Kara ()
1. Introduction
Surgery is a branch of medicine that involves manual and instrumental intervention. Globally 313 million surgical procedures are performed each year and approximately 4.2 million postoperative deaths occur each year, accounting for 7.7% of global mortality. Events occur during and after surgery that constitute intraoperative and postoperative complications, including death. A death occurring after surgery is called postoperative death. It can be early or immediate if it occurs within forty-eight (48) hours and late if it occurs within thirty (30) days [1] .
Surgery is an integral part of effective health systems, with surgical diseases accounting for 30% of the global burden of disease [2] .
Postoperative mortality is one of six surgical indicators identified by the Lancet Commission on Global Surgery to monitor access to high-quality surgical care [3] .
Postoperative mortality is the oldest form of criticism and evaluation in surgery [4] , and in an intensive care unit.
In a hospital department, such a study allows a control and review of therapeutic measures; these are likely to deteriorate in their implementation over the years, which requires periodic criticism [5] .
Despite constant progress in surgery and anesthesia, the mortality rate related to surgery in industrialized countries is estimated at 0.4% to 0.8%, and the rate of serious post-operative complications at 3% to 16%. In developing countries, the mortality rate is 5% to 10%. Each year, approximately seven million patients suffer from surgical complications, of which at least one million die during or immediately after surgery [6] .
Very few studies have been carried out in Togo on this subject, but some have shown a rate of 15.3% in 2006 [6] .
What about early postoperative deaths in the surgical intensive care unit at CHU-Kara? To answer this question, we proposed to take stock of early postoperative deaths in the surgical intensive care unit of this hospital with the specific objectives of
- to identify the cases of early postoperative deaths;
- to identify the risk factors for pre, per and postoperative death;
- to analyze the risk factors.
2. Methodology
Our study took place in the surgical resuscitation department and operating room of the University Hospital of Kara (CHU-Kara), a national reference center in the northern part of Togo covering one third of the Togolese population. It has a capacity of 204 beds.
The surgical rehabilitation had a capacity of 8 beds.
It was equipped with two multiparametric monitors, one of which was functional, a functional respirator, a syringe pump, a saturometer, emergency drugs, and a laryngoscope.
One anesthesiologist resuscitator (and for a third of the population of Togo), one anesthesiologist paramedic; four (4) nurses; four (4) patient guards ensured the activities in surgical resuscitation. A team composed of a nurse, a paramedical anesthetist serving both the operating room and the resuscitation covered during the same period the activities of the surgical resuscitation and a patient guard every day including weekends and holidays.
There were three (3) operating rooms. The anesthesiologist resuscitator (head of anesthesia and resuscitation); surgeons (2 visceral surgeons, a neurosurgeon, a pediatric surgeon, three urological surgeons, an ophthalmologist, two ENT, two traumatologists, an obstetrician gynecologist); six (6) nurse anesthetists; five (5) instrumentalists; six (6) nurses; two (2) DES of general surgery and one DES of gynecology and obstetrics provided anesthesia and surgery.
This was a retrospective chart review of patients who died in early postoperative care from November 1, 2019 to April 30, 202 (18 months). Data collection was performed from May 11 to October 12, 2021 (06 months).
Included were the records of all patients operated in emergency or program, and died within 48 hours after surgery during the study period in surgical resuscitation.
Excluded were the records of patients operated in emergency, or in program, who died intraoperatively, or postoperatively in the ICU after 48 hours, or outside the ICU.
The means used were medical records, therapeutic and monitoring sheets, anesthesia records, surgical report registers, surgical resuscitation registers, and anesthesia registers of the operating room.
A self-developed data collection support form was used (see Appendix). It is not a questionnaire.
The data were collected from medical and anesthetic records, treatment sheets, operative report registers, and anesthesia sheets. The supervisors of the surgical resuscitation and operating room departments were questioned about the structure, human and material resources, and work organization.
The parameters studied were:
- Preoperative risk factors: sociodemographic and epidemiological data; reason for and delay in consultation or admission; clinical status of the patient, ASA classification; personal medical, obstetrical and surgical history, family history, current treatments.
- Intraoperative risk factors: surgical lesion; procedure performed, emergency or scheduled surgery, intraoperative adverse events, time to surgical treatment, duration of surgery, type of anesthesia and anesthetic drugs used, intraoperative complications.
Early postoperative risk factors: postoperative complications, favouring factors (patient-related, surgical and anaesthetic).
- Obvious causes of death.
The survey form was used to collect data.
The main criterion was mortality within 48 hours after surgery.
The qualitative data were expressed as numbers and percentages.
For the study to be carried out, a written request for authorization was submitted by the research team and countersigned by the team leader to the director of the CHU-Kara. This request received a favorable opinion from the director, which allowed the work to begin on May 11, 2021.
3. Results
Sociodemographic and Epidemiological Aspects
Number patients: Thirty-two early postoperative deaths or 2.22% out of 1442 operated with 30 deaths retained for the study.
Sex of deaths: There were 21 (70%) male and 9 (30%) female deaths with a sex ratio of 2.33.
1) Ages of deceased patients
The mean age was 31.22 years with the extremes of D1 and 75 years.
D1 corresponds to one (1) day of life.
2) Profession of deceased patients
3) Time to surgery
The average time to surgery was 1.6 days with extremes of 0 and 8 days.
4) Clinical aspects
a) History and associated pathologies of the deceased patients
Two patients were hypertensive, one patient was operated for acute appendicitis.
b) Pathologies of the deceased children
Two children had omphalocele, one child had Hirschprung's disease. Acute intestinal intussusception was diagnosed in three children. Anorectal malformation was diagnosed in one child.
5) Physical condition of the deceased patients: ASA score
6) Surgical aspects
a) Circumstances of surgery
Twenty-four (80%) patients were operated on as emergencies and 6 patients (20%) were operated on as a program.
b) Intraoperative surgical lesions
7) Duration of surgery
The average duration was 132.5 minutes with extremes of 60 and 240 minutes.
8) Anesthesiological aspects
a) Types of anesthesia
General anaesthesia was used in 26 patients (86.66%) and 4 patients (13.34%) received locoregional anaesthesia.
b) Anesthetic drugs used
9) Complications
a) Intraoperative complications
b) Postoperative complications
10) Risk Factors
a) Risk factors related to the patients’ past history and associated pathologies
b) Risk factors for infection
c) Surgical risk factors
d) Anesthetic risk factors
e) Obvious causes of postoperative death
f) Correlation between preoperative risk factors
The relationship between age and intervention delay is not significant.
The relationship between gender and intervention delay is not significant.
The relationship between farmer and intervention delay is significant.
g) Correlation between intraoperative risk factors
Digestive surgery had a correlation with mode of surgery.
General anesthesia had a correlation with mode of surgery.
The duration of the operation did not correlate with the mode of operation.
h) Correlation between postoperative risk factors
Dirty and contaminated surgery correlated with postoperative complications.
General anesthesia correlated with postoperative complications.
Duration of surgery correlated with postoperative complications.
Intraoperative complications correlated with postoperative complications.
4. Discussion
Our discussion focused on:
Methodology (limitations or difficulties of the study);
Sociodemographic and epidemiological aspects;
Clinical aspects; surgical aspects;
Complications; risk factors; obvious causes of death;
Correlation between preoperative risk factors;
Correlation between intraoperative risk factors;
Correlation between postoperative risk factors.
5. Methodology (Limitations)
The retrospective descriptive study focused on early postoperative deaths in the surgical intensive care unit at CHU-Kara during an 18-month period.
The difficulties were related to the retrospective study: by the inexistence of a death register; in 15 files it was impossible to collect certain data (hence the notion of not informed). Two files were not usable and were not taken into account.
5.1. Sociodemographic and Epidemiological Aspects
Number of patients: thirty-two (32) deaths occurred out of 1442 operated on in 18 months with an early postoperative mortality rate of 2.22%.
This rate seems very high in our context, depending on the standard of living of the population, the inadequacy of the technical facilities and the non-medicalization of anesthesia, which was noted since 2008 by Ouro-Bang’na [7] in Togo with a rate of 0.89%. There is a clear difference at 420 km from Lomé (in Kara).
Hajanindrina [8] in Madagascar and Mallol [9] in Spain found a respective rate of 1.6% in 2016 and 0.8% in 2013. The result of Hajanindrina [8] was lower but still characterized by the phenomenon of underdevelopment while Mallol [9] in Spain was lower with the characteristics of high means countries.
According to Assouto [10] , the under-equipment of health facilities and the glaring absence of qualified personnel (almost non-existence of anesthesiologist in our context) constitute a real handicap for the management of anesthesiology.
According to Meara [11] , remarkable progress has been made in global health over the last 25 years, but this progress has not been uniform. Mortality and morbidity from common conditions requiring surgery have increased in the poorest countries or regions of the world. At the same time, the development of safe, essential, life-saving surgical and anesthetic care in low- and middle-income countries has stagnated or regressed. In the absence of surgical care, lethality and mortality rates remain high for common and easily treated conditions.
Male patients were the most affected by early postoperative mortality with a sex ratio of 2.33. This result would be related to the number of male patients operated on, which was more numerous than the female sex contrary to that of Ouro-Bang’Na [7] in 2008.
This result is identical to that of Serji [4] in Morocco in 2010. In Spain with Mallol [9] and in Madagascar with Hajanindrina [8] the male sex was more represented respectively (68%, 72.13%). Women generally use health care services more and faster than men.
The average age was 31.22 years and the age group [16 - 30 years] had 10 patients (30%) (See Table 1). This predominance of young people is related to the young population in our developing country context. On the other hand, Mallol [9] in Spain in 2013 found the age range [61 - 81] predominant. This older population is characteristic of developed countries.
In Africa, patients are younger and undergo less extensive procedures. In theory, mortality should therefore be lower than elsewhere, but according to Massamba Diop [12] , the opposite is true, in particular because of postoperative complications dominated more by postoperative infections and by delays in
![]()
Table 1. Age distribution of deceased patients.
consultation and treatment.
Farmers were the most affected by mortality with a rate of 66.67% (Table 2). This result would be related to the fact that the population of the city of Kara and its surroundings was dominated by peasants who did not respect hygiene and disease prevention measures. Results found by Koffi [13] in Côte d'Ivoire were similar with a rate of 51.4%. This predominance of farmers with a low socio-economic level and the absence of health coverage favored delays in consultation and treatment. The extension of health insurance to this socio-professional class, which is underway in our context, will reduce deaths related to delays in consultation and treatment.
The mean delay to surgery was 1.6 days (Table 3). This delay was long. Insufficient financial means to honor the prescriptions was one of the factors prolonging this delay. On the other hand, according to Tavernier B [14] , in developed countries, emergency surgeries were performed after admission, followed by management with a short delay because there is functional universal health coverage (especially in emergency).
5.2. Clinical Aspects
1) Congenital malformation of deceased patients
42.85% of the children had a malformation that was a risk factor. The causes of these malformations were not well known.
2) Physical condition of the patient: ASA score (Table 4)
Patients with an ASA I score were in the majority with a rate of 40%. These patients were probably poorly evaluated or poorly managed intra- and postoperatively. This refers to the non-medicalization of anesthesia in our context where there was only one anesthesiologist to supervise the anesthesia 24 hours a day, 7 days a week. For the other patients, the risk was higher (ASA > 2) and would better explain these deaths. Fecho [15] in 2008 found that the majority of patients who died in early postoperative care were ASA V. Mallol [9] on the other hand found 84% of patients to be ASA II.
The higher the ASA score, the greater the risk of complications and death.
3) Surgical aspects
Urgent surgery represented 80% and was a high risk factor for complications.
![]()
Table 2. Distribution of deceased patients according to profession.
![]()
Table 3. Distribution of deceased patients according to time to surgery.
![]()
Table 4. Distribution of deceased patients according to ASA score.
Hajanirina and Assouto [8] [10] found higher rates, respectively 86% and 89.2%. This is characteristic of the surgical management in developing countries. Contrary to studies done in developed countries, with a lower rate, Fecho K in the USA, Tavernier in France and Mallol in Spain [9] [14] [16] respectively 30%, 42% and 49.9%.
Mortality was particularly high in patients operated on in emergency, proven by the literature Meara and Koffi [11] [13] . This difference in rates between developing and developed countries can be explained by the rapid and efficient management of the patient, with a high performance technical platform and a sufficient number of high performance personnel, which was not the case in our context where the means and conditions were deficient and the anaesthesia was not medicalized according to Ouro-Bang’Na [7] .
4) According to the type of lesion
Digestive surgical pathologies were diagnosed in 76.66% of the deceased patients (Table 5).
Surgical emergencies of the gastrointestinal tract had accounted for a significant share of surgical pathologies. According to Sama [17] in 2017 in Togo, 59% of emergencies were abdominal emergencies.
![]()
Table 5. Distribution of patients who died according to the intraoperative surgical lesion.
Digestive surgical pathologies have a very high frequency of mortality due to hypovolemic and septic shock.
5) Surgical treatment
Digestive surgery predominated the table (76.66%) (Table 6).
These surgeries were the source of many complications leading to immediate postoperative death.
This result is almost identical to the data collected by Tavernier and Mallol [9] [14] respectively 89.58% and 76.25%.
Digestive surgeries were highly infectious dirty surgeries with a significant risk of complications and death.
6) Duration of surgery (Table 7)
The average duration was 132.5 minutes in our study, this long duration would be related to complex surgeries or surgeries performed by the learners
![]()
Table 6. Distribution of patients according to the duration of the operation.
![]()
Table 7. Distribution of patients according to the anesthetic drugs used.
(junior).
Our result is similar to that of Bonkoungou [18] who reported an average duration of 125.62 minutes.
Harouna [19] , in his study on intestinal obstructions, noted an average operating time of 3 hours in 92% of postoperative deaths.
This long duration can be a factor in complications and mortality. The longer the duration of the procedure, the more the risks associated with anesthesia synergize with the surgical aggression. Short surgery in patients in poor condition would decrease morbimortality.
5.3. Anesthesiological Aspects
1) Type of anesthesia
General anaesthesia was performed in 86.66% of the deceased patients, it was the most indicated according to the surgical lesions presented by the deceased patients (digestive pathologies).
These results were close to those of Hajanirina [3] who found 79.1%. This anesthesia was known to be associated with cardiovascular, respiratory and neurological complications.
2) Anesthetic drugs used
The protocol associating the drugs Propofol + Fentanyl + Ketamine + Atropine was more used for general anesthesia (Table 8). Given the surgical lesions, the clinical condition of the patients, and the accessibility of the above-mentioned products, the products most indicated were the least aggressive and the least toxic.
3) Intraoperative complications
They were represented by intraoperative cardiac arrest and hemorrhage in 20% of cases (Table 9). This result would be related to the advanced pathology and/or the complexity of the surgical lesion or the incompetence of the actors and/or the side effects of the anesthetic drugs. These intraoperative complications were the significant risk in the occurrence of postoperative complications (82.30%).
4) Postoperative complications
They were dominated by medical complications (46.66%) (Table 10), represented by cardiovascular complications. The extent of the lesions and the significant blood loss during the operations could explain these results. Complications in our context were lower than that of Fecho [16] with 85%.
Postoperative complications were significantly correlated with the duration of the operation (57.90%), general anaesthesia (84.60%); contaminated and dirty surgery (76.5%).
![]()
Table 8. Distribution of patients according to intraoperative complications.
![]()
Table 9. Distribution of patients according to postoperative complications.
![]()
Table 10. Distribution of patients according to patient-related risk factors.
There was a clear evolution in the management between 2002 and 2006 according to Ouro-Bang’Na [7] but the interior of Togo is at a disadvantage with the example of the Kara University Hospital with 2.22% of early postoperative mortality.
6. Risk Factors
1) Patient-related risk factors (Table 11)
Half (50%) of the patients who died had consulted late and were therefore in the advanced stages of their disease. The low socioeconomic level of the population and the delay in treatment were risk factors for mortality in our context as in developing countries.
2) Surgical risk factors
Dirty surgery was performed in 46.67% of cases (Table 12). It was associated with urgent, complex and/or advanced digestive surgical pathologies aggravating the risk factors. Intraoperative hemorrhage was the major risk factor.
3) Anesthetic risk factors (Table 13)
Delayed awakening was the most frequent factor in the patients who died, with a proportion of 16.66%, and general anaesthesia favoured postoperative complications in 84.60% of cases.
![]()
Table 11. Distribution of patients according to Altémeier class.
![]()
Table 12. Distribution of patients by surgical risk factors.
![]()
Table 13. Distribution of patients according to anesthetic factors.
Anesthesia had a correlation with the mode of operation and contributed to the occurrence of postoperative mortality in our context.
4) Obvious causes of postoperative deaths (Table 14)
The etiologies of death were more of a cardiovascular nature with a proportion of 20%. For Koffi, Yenon, Hajanirina and Tavernier [8] [13] [14] [20] infectious etiologies were the most represented with respectively 43.05%; 40%; 54.54% and 42%.
This disparity is explained by the fact that septic shock is the first cause of postoperative death in these countries, and the physical condition and age of the patients favored septic shock.
5) Correlation between preoperative risk factors
Twenty (66.67%) patients were the farmers, had consulted late (Table 15). Ten patients (33.33%) had consulted early.
There was a statistically significant relationship between occupation and delay
![]()
Table 14. Distribution of patients by causes of postoperative death.
![]()
Table 15. Correlation between socio-demographic data and intervention delay.
in consultation (p = 0.00001).
Poverty, self-medication and ignorance of serious diseases were the factors explaining this delay in consultation by the peasants.
6) Correlation between intraoperative risk factors
Twenty-three (76.70%) patients with digestive surgical pathologies were operated on as emergencies (Table 16). There was a statistically significant correlation between digestive surgery and urgent intervention (p = 0.04231). The emergency with its factors combine with the unfavorable hemodynamic factors to create more aggravation of the postoperative evolution.
General anesthesia was performed in 26 (86.70%) patients in emergency.
There was a statistically significant correlation between the circumstances of intervention and the type of anesthesia (p = 0.03884). General anesthesia offers a similar situation of unfavorable hemodynamics and emergency factors to accentuate the poor postoperative course.
Perioperative management of abdominal surgical emergencies remains a major concern due to intraoperative hemodynamic and metabolic disorders, but also postoperative complications.
![]()
Table 16. Correlation between intraoperative risk factors and mode of operation.
![]()
Table 17. Correlation between risk factors and postoperative complications.
7) Correlation between postoperative risk factors
Fourteen (46.66%) patients had undergone dirty surgery and had experienced cardiovascular complications and delayed recovery postoperatively (Table 17). Twelve patients (40%) had undergone contaminated clean surgery.
There was a statistically significant correlation between dirty surgery and postoperative complications (p = 0.000891).
Intraoperative hemodynamic instability, general anesthesia and its cardiovascular complications had favored deaths.
7. Conclusions
The study had difficulties and limitations for its realization, inherent to a retrospective study.
At the end of the study we retained:
The early postoperative mortality rate was 2.22%. The majority of patients were male. Most of them were young.
Low socioeconomic status, delayed consultation, patient condition, and congenital malformations were preoperative risk factors.
The long delay of intervention, digestive surgical pathologies, and surgeries performed in emergency, general anesthesia, the occurrence of intraoperative cardiac arrest, and the non-medicalization of the anesthesia were intraoperative risk factors.
Cardiovascular complications, delayed recovery, dirty or infected surgery were the postoperative risk factors.
Cardiovascular damage was the additional cause of death in the majority of cases.
Intraoperative complications were high, as were very high postoperative complications.
Anesthesia and surgery, additional intraoperative factors; as well as the low level of qualification of the intraoperative actors, the lower intraoperative security; the combination of anesthetic effects, the complexity of intraoperative lesions and the level of management due to the low level of the actors led to the increase of this mortality.
The combination of two or more factors was pejorative for the majority of deaths.
Parameters were missing to show the epidemiological aspect, but these results encourage further work to improve the therapeutic principles.
Once again, the study sheds light on the socioeconomic factors (poverty), the pre, per and postoperative risk factors.
Appendix. Survey Sheet
SURVEY SHEET No........................................................
- Marital status
Name...................................... First name......................................
Age................ Sex F................ M ................
Occupation...................................
Reason and date of admission....................................................
- Exam
Clinical condition of the day.........................................................
- History
Personal
Medical: HTA ….Diabetes…. Sickle cell disease….. Epilepsy……….
Asthma ................Other (specify) ................
Surgical....................................................................................
Obstetrical:
Birth parameters (child)… Resuscitation…Malformation …Other (specify)...
Current treatments: Corticosteroid therapy…. Chemotherapy……
Radiotherapy and Other (specify)................................................
Familial: Specify family defects................
- Surgery
Type of lesion....................................
Procedure performed and date................
Emergency ................Scheduled surgery................
Duration of surgery (minutes)…
- Anaesthesia
Type of anaesthesia: General................Locoregional (Specify) ................
Drugs used............................................................................
- Complications
Intraoperative anaesthetic complications: hypoxia….Anoxia…. cardiac arrest…
Hypothermia................ Other................
Post-operative complications: delayed awakening….Shock…….
Intra-operative cardiac arrest................ Other................
- Deaths
Contributory factors
Pathological...
Surgical...
Anaesthetics...
Obvious aetiologies: coagulation disorder…advanced stage of disease……… haemorrhage………..
Other (specify).............................................................................................