Hystererectomy in a Tertiary Hospital in a Sub-Saharan Setting: A 20-Year Retrospective Review of the Indications, Types and Analysis of Technical Index ()
1. Introduction
Hysterectomy is amongst the most performed surgical interventions in Gynaecology [1]. This surgical intervention is usually done for benign pathologies like uterine fibroids and is reported to represent up to 20% of gynaecological interventions in some countries [1] [2]. The main indications of hysterectomies reported in the literature are benign pathologies of uterus like leiomyomas [1] [2] [3]. For a long time, abdominal hysterectomy has remained the main type followed by vaginal route especially in developing countries. Buambo et al. in a retrospective study done in Brazzaville found that abdominal route was the main route (82.5%) followed by vaginal route (17.5%) [3]. Development and popularisation of minimally invasive approaches and alternatives like endometrectomy in recent years have enabled the reduction of complications, morbidity and hospital stay linked to abdominal approach. The trend in the recent years, in developed countries, has been to use minimally invasive techniques both for benign and malignant uterine diseases [4] [5]. The first laparoscopic hysterectomy is reported to have been performed in 1989 and demonstrated several advantages as compared to the traditional abdominal route like less post-operative pains, better cosmetics, shorter hospital stay, early recovery time [4] [5] [6] and [7]. Laparoscopic and vaginal routes are now the standard over the abdominal route whenever possible in western countries [4] [5] as compared to Subsaharan settings where vaginal and laparoscopic routes are seldom performed [3]. Many authors think that the technical index of hysterectomy is an indicator of the quality of gynaecological service. Mindful of this, in a bit to improve in our practice, we designed this study to audit the practice of hysterectomies for obstetrical and gynaecological reasons in our service with regards to the frequency, the indications, the types and to take stock of our technical index.
2. Material and Methods
2.1. Study Design, Period and Site
We carried out a cross-sectional retrospective study, in the department of Obstetrics and Gynaecology of the Douala General Hospital (DGH), from the 1st January 2000 to 31st December 2019, in order to assess the frequency, indications, surgical methods and complications of hysterectomy.
The DGH is a tertiary health structure in Douala, the largest city and economical capital of Cameroon, with an estimated population of over three million inhabitants. This hospital has a capacity of 320 beds and hosts many services, amongst which is the service of Obstetrics and Gynaecology. The workforce of the Gynecology and Obstetrics department is constituted of 8 gynaecologist and Obstetricians, with few having special training in vaginal surgery, Laparoscopy and other surgical techniques. The service is involved in the training of medical students and resident doctors in Obstetric and Gynecology.
2.2. Inclusion and Exclusion Criteria
Patients included were those who have undergone hysterectomy for gynaecological or obstetrical indications, in the service of Obstetrics and Gynaecology of the DGH, and whose files were complete. A file was considered complete if it contains beside others at least the following information: age, the indication, the route of surgery, the duration of hospital stay, and the outcome of management. Files with more than 25% (more than one these key data) of missing data were excluded from the study.
2.3. Data Collection
A list of patients who underwent hysterectomy was obtained from the theatre and inpatients registers. The files of the patients were then compiled and information was retrieved and filled in a pre-established data collection form. The data recorded included the age, surgical indication, surgical method, intra and post-operative complications and the length of hospital stay. Blood loss of more than 500 ml was considered as hemorrhagic complications. Febrile morbidity was considered for a postoperative temperature of above 38˚C.
2.4. Statistical Analysis of Data
The data were typed and analysed using the software SPSS 21.0 version. The results were reported in terms of number, mean, and percentage. The differences between the groups were determined with the Fisher’s test with p ≤ 0.05.
2.5. Ethical Considerations
The study was endorsed by the national ethical committee. Access to files was granted only after approval from the ethical committee of the institution. The data were exploited discreetly.
3. Results
968 files fulfilled inclusion criteria and were enrolled in this study. The sociodemographic characteristics, the frequency, the indications of the surgery, the route of surgery, the hospital stay, and the complications are summarized in Table 1.
3.1. Age
The mean age of the patients was 45.75 ± 7.71 years (extremes 19 to 75 years). The age range of 40 to 50 years was the most represented 56.40%: (546 cases) (Table 2).
3.2. Frequency
Out of 7126 surgical cases realised in the service, 1007 patients underwent an
Table 1. Frequency of hysterectomies at Douala General Hospital from 2000 to 2019.
1007 cases of hysterectomies were performed. It represented 14.21% of surgical operations, hence an average of 43.8 cases per year.
Table 2. Distribution of hysterectomies by age.
The mean age of the patients was 45.76 ± 7.71 years with a minimum of 19 years and a maximum of 75 years. Patients over 40 years old were the most represented.
hysterectomy, giving a frequency of 14.21%. Thirty-nine incomplete files were excluded and our study involved 968 patients.
3.3. Indications
The indications included symptomatic fibroids 64.15% (621 cases), gynaecological cancers 13.94% (135 cases), severe cervical dysplasia 11.15% (108 cases), and endometrial hyperplasia with atypia 7.02% (68 cases), haemostatic hysterectomies 2, 68% (26 cases), uterine prolapse 0.82% (8 cases), one case of post abortion uterine necrosis (0.10%) and one case of uterine endometriosis (0.10%) (Table 3). Total hysterectomy was performed during debulking surgery in 76 cases of ovarian cancers while 53 cases included expanded hysterectomies for cervical cancer. Subtotal hysterectomy was performed for 1 case of post abortion uterine necrosis and in 4 cases of debulking surgery for frozen pelvis.
3.4. Surgical Method
Laparotomy was the main surgical type 85.95% (832 cases), followed by the vaginal route 10.20% (97 cases) and laparoscopy 3.92% (38 cases). All the laparoscopic hysterectomies were total, and involved benign uterine pathologies in patients without a history of prior abdomino-pelvic surgery and in whom the uterine size did not exceed 12 weeks of amenorrhea. Fourty percent of laparoscopic hysterectomies performed were type II according to “classification of Mage” [2] (Table 4).
3.5. Duration of Hospital Stay
The average length of hospital stay was 6.07 ± 1.92 days following laparotomy, 3 ± 1.09 days following the vaginal route, and 3.6 ± 1.04 days following laparoscopy. Of the three surgical methods, the length of hospital stay was longer following abdominal hysterectomy with a significant difference (p = 0.00023) (Table 5).
3.6. Complications
We found a global rate of complications of 11% (106 cases), with 3.5% (34 cases) intra-operatively and 7.5% (72 cases) postoperatively (Table 6).
Intra-operative complications included haemorrhage, 1.75% (17 cases), bladder
Table 3. Distribution of hysterectomies according to indications (n = 968).
Symptomatic myomas, severe cervical dysplasia and cervical cancer were the main indications in order of importance.
Table 4. Hysterectomies and surgical methods.
Laparotomy was the main route (86.05%) of hysterectomies followed by vaginal route (10.02%) and laparoscopic route (3.92%).
injuries 0.82% (8 cases), ureteral injuries 0.72% (7 cases) digestive wound 0.10% (1 case) and death 0.10% (1 case) (Table 6). Postoperative complications included: fever 3.61% (35 cases), anemia 2.5% (24 cases), abdominal wall sepsis 0.92% (9 cases), hemoperitoneum 0.10% (1 case) and death 0.10% (1 case). Intra and post operative complications observed were significantly higher following laparotomy than the vaginal route and laparoscopy (p = 0.003) (Table 7).
4. Discussion
4.1. Age
Our mean age is comparable to that reported by Bamboet al. in Congo (42.7 years), Razafindrabe et al. in Madagascar (42.5 years) and Kouam et al. in Cameroon (43.2 years) [3] [6] and [7]. Like Bambo et al. in Congo [3], our population is still influenced by herbal medicine, even in cases of symptomatic fibroid [8] [9] [10] and [11]. This can last for several years, therefore delaying the surgical treatment of benign lesions like uterine myomas at age of 40 or more.
4.2. Frequency
Matanga et al. and Egbe et al. reported a frequency of hysterectomy of 14.54% at the DGH in 2010 [9] [12]. This implies that the practice of hysterectomy has
Table 5. Surgical methods and Hospital Stay.
Mean hospital stay was 6.07 ± 1.92 days with a minimum of 3 and a maximum of 10 days. The majority of patients (75.6%) spent 5 to 7 days postoperatively. Of the first three routes, hospital stay was longer for abdominal hysterectomy with a significant difference (p = 0.00023).
Table 6. Surgical methods and intraoperative complications (n = 34).
Hemorrhage (12) was the most common cause of intraoperative complications. Followed by vesical wounds (1) and digestive wounds. And these complications are most found when using the high way.
Table 7. Surgical methods and postoperative complications (n = 72).
Fever and anemia were the postoperative complication found in the three pathways. Postoperative complications after laparotomy were greater than those observed vaginally and laparoscopically significantly (p = 0.003).
remained constant, in the service of Obstetrics and Gynecology of the DGH. Our frequency is higher than that found at CHU of Yaoundé, 9.33% and CHU of Brazzaville, 6.6% [3] and [7].
4.3. Indications
The main indications of the hysterectomies were symptomatic fibroids in 64.15% of cases. This rate is close to 64.86% found by Kouam et al. and 63.9% of Buambo et al. [3] [7]. This is in phase with literature, which ranks symptomatic uterine fibroids as the principal indication of hysterectomy [10] [11] [13] [14], and [15]. A study carried out by Leveque et al. in France, had revealed the predominance of the benign uterine pathology in the order of 70% amongst the gynaecological indications of hysterectomies [16]. The high rate of fibroids in our study is justified by the high prevalence of this pathology in the female black race [6] [17].
Obstetrical indications represented 2.62% of cases of hysterectomies in our Hospital and were mainly done for life saving hemostatic reasons.
4.4. Surgical Methods
The analysis of postoperative notes identified 234 cases (33.19%) of abdominal hysterectomies which could have benefited from the vaginal route or laparoscopy. Presently, most data strongly suggest a strong morbidity linked to abdominal hysterectomy which is not the case for minimally invasive techniques which include the vaginal routes and laparoscopies [18] [19] [20] and [21]. Consequently, the tendency now is preference of minimally invasive techniques of hysterectomy, over abdominal hysterectomy [20] in developed countries. We attribute this low index of laparoscopic surgery to the fact that not all the gynaecologists are trained to the technique in one hand and the members of the team like theatre nurses, anaesthetists trained to laparoscopy are lacking. The calculated technical index of hysterectomy in our service was 12.9% in 2017 [21] close to the 19% found by gynaecologists of the service of Maha Alkhadury et al. in the Middle East [22] [23], lower than 30% - 60% found by Laberge and Singh in Quebec [24] and 90% in France [25].
4.5. Length of Hospital Stay
Our lengths of hospital stay are short for the vaginal routes and laparoscopy. Thanks to a good learning curve and a better control of post-operative pain, the vaginal and laparoscopic methods can reduce the hospital stay to shorter than 24 hours and consequently: reduce the costs of hospitalisation, favouring a rapid healing and early recovery of the patient’s societal activities [22] [26] [27] and [28].
4.6. Complications
Hemorrhagic complications in our study are explained by hysterectomies done on large polymyomatous uterus, likely in the presence of adhesions, as sequelae of chronic pelvic inflammatory disease [29]. Half of the urinary injuries occurred during debulking surgery for ovarian cancer and during radical hysterectomies and lmphadenectomy for cervical cancer. Post-laparotomy fevers are explained essentially by the large cruentous surfaces, necrosed tissues and residual bleeding (which is less likely with minimally invasive surgery), sources of release of pyrogens [22] [30] [31]. The maternal death observed in our study was due to a complication of septic abortion, despite the successful hysterectomy. Our complications are similar to those observed in literature [3] [7] [31] and [32].
5. Conclusion
The frequency of hysterectomy is 14.21%. Uterine fibroids, gynaecological cancer and cervical dysplasia are the main indications. Intra-operative haemorrhage, bladder and ureteral injuries are the major complications. There is a longer hospital stay and a higher rate of complications following laparotomy than minimally invasive methods, in our context where the technical index of hysterectomies is low. The increase in the practice of the vaginal and laparoscopic methods through reinforcement of abilities could contribute to reducing the peri- and post-operative complication as well as hospital stay.
6. Study Limitations
This study was retrospective with 39 out of 1007 files not included because of missing data; this can create an important bias in the analysis of data.
Furthermore, other important aspects of the technical index like the specific training of gynaecologists in areas like minimally invasive surgery, vaginal hysterectomy or other conservative surgical management of postpartum haemorrhage couldn’t be assessed.
Authors’ Contribution
Théophile Nana Njamen and Robert Tchounzou designed the study and wrote the manuscript. Fulbert Nkwele Mangala, Adamo Bongoe, Fidelia Mbi Kobenge participated in patients’ recruitment and manuscript revision. All the other authors revised the manuscript.
All the authors read and approved the final version of the manuscript.
Appendix
XI-1 Data collection form
Hysterectomy in a tertiary hospital in a sub-Saharan setting: A 20-years retrospective review of the indications, types and analysis of technical index
Data collection sheet
I. Identification
a) Name or ID:__________________________DATE: ____/_____/___________
b) Age (years): ________ c/ Parity: _______ d/ Residence: ________________
c) Level of education: i/ Primary ii/ Secondary iii/ University
d) Profession: i) NE ii) Self-employed iii) Civil servant iv) Private salary v) Student
II. Past history
a) Chronic disease: i) Sickle cell anaemia ii) Diabetes iii) HTN (hypertension) iv) Cancer v) other
b) Past surgery: i) No ii) Yes Indication and site_________________________
c) Tobacco consumption: i) No ii) Yes
d) Chronic infection: i) HIV ii) Tuberculosis iii) Hepatitis
III. Indications of Hysterectomy (stick all correct options)
a) Uterine myomas
b) Uterine prolapse
c) Uterine rupture
d) Unexplained abnormal uterine bleeding
e) Ovarian cancer
f) Cervical dysplasia
g) Cervical cancer
h) Endometrial cancer:
i) Endometrial atypic hyperplasia
j) Endometrial cancer
k) Post-partum haemorrhage (indicate the cause)
IV. Type of hystetrectomy
a) Abdominal hysterectomy
b) Vaginal hysterectomy
c) Laparoscopic hysterectomy
V. Per operatory events
a) Type of anaesthesia: i) Local ii) Spinal iii) epidural iv) General
b) Duration of operation: i) less than 30 min ii) 30 - 60 min iii) 60-120 min iv) > 120 min
c) Per-op complications: i) bowel injury ii) Bladder injury iii) haemorrhage iv) Ureteral injury
v) other________
VI. Post-operative events
a) Duration of hospital stay___________________
b) Haemorrhage
c) Peritonitis
d) Urinary tract infection
e) Surgical site infection
f) Phlebitis
g) Fever
h) Pulmonary embolus
i) Vesico-vaginal fistula