It should be noted that the prevalence of infertility is higher in Africa than in the countries of the North. The infertility rate in sub-Saharan Africa is even one of the highest in the world, and 15% to 30% of couples are affected by this problem compared to 5% to 10% of couples in developed countries [4] [5]. Epidemiological studies carried out in Senegal estimated the couple’s infertility in Senegal between 12% and 14%, in young women (average age 32 years) with a tubo-peritoneal cause which was found in 21% to 61% [6] [7] [8]. Thus, we can understand that with this etiological profile of female infertility, ultrasound and HSG are the first-line examinations to search for pelvic anatomical pathology. All of our patients had undergone HSG for this purpose. In developing countries like Senegal, mechanical causes by tubal obstruction are thought to play a determining role in the onset of infertility. These causes are certainly linked to the high prevalence of infections for which the management would not be adequate.

In our series, 43% of the tubes blocked at HSG were permeable at laparoscopy. In the series by Mol et al. [9], the same observation was found in 40% of proximal obstructions. A study carried out in France [10] showed in HSG, a rate of tubal spasm of 40% which was revealed to be permeable to laparoscopy. The existence of a tubal spasm proximal to HSG therefore justifies performing a laparoscopy in order to rule out or confirm the diagnosis. The presence of false positives for HSG can be explained by spasms in response to pain and mucous plugging. Certain measures before performing HSG make it possible to reduce the rate of false positives, such as the use of analgesics, psychological preparation, traction of the uterine cervix in order to reduce possible anteversion or retroversion and above all a good interpretation of the pictures [3]. The latter presents a significant inter-operator variability, estimated at more than 20% [11]. It should be recognized that good reliability of HSG in the diagnosis of proximal tubal obstructions would render laparoscopy unnecessary and would justify a switch to in vitro fertilization [3] [12]. Regarding tubes permeable to HSG and obstructed by laparoscopy, some authors recommended a deep anesthesia before reinjecting the bruise. This device allowed them to objectify a tubal passage 3 times out of 4 [3]. Tshabu [13] in Benin uses this device to relieve the tubal spasm. Another explanation for the tubal obstruction found on laparoscopy would be the long time that elapses between performing HSG and laparoscopy, which may explain the appearance of tubal pathology at laparoscopy although it is absent at HSG.

Distal tubal obstructions are accessible to therapeutic surgical procedures [14]. This diagnosis justifies the practice of a laparoscopy for therapeutic purposes. In our study, the 2 examinations were in agreement concerning the distal tubal patency in 62% of cases with a Kappa test at 0.40 indicating a moderate correlation. These rates are found by Tsabu [13] (67%) and Kehila [3] in 83.3% with a k = 0.53. In fact, the difficulty in this case is above all to differentiate bilateral tubal patency from unilateral distal obstruction at HSG. Because the vision of the tube injected up to its distal part associated with the peritoneal mixing from the permeable tube can easily be confused with bilateral patency [3].

The relationship between pelvic adhesions and female infertility is recognized. These adhesions alter the delicate anatomical relationship between the tubal bulb and the ovary, interfering with or preventing normal capture of the oocyte [15]. Its prevalence in infertility is estimated between 10% and 23% [9]. The reliability of HSG in peritoneal assessment is far from absolute. Most of the studies argue in favor of the clear superiority of laparoscopy in this indication. We believe that, providing a direct view of the pelvis, laparoscopy can be considered the gold standard for detecting pelvic adhesions.

The sensitivity of HSG in detecting pelvic-peritoneal adhesions was very low (4%) with a K = −0.008. This rate is lower than that found by Kehila [3] 9.5%. In Dadoun’s study [10], 80% of patients had adhesions found on laparoscopy. Ngowa [16] found sensitivity (24.6%) of HSG in the diagnosis of adhesions. In summary, for the assessment of adhesions, the HSG was unreliable [17] [18].

Given the high frequency of tuboperitoneal lesions in Africa, early management of sexually transmitted infections could certainly prevent tubal infertility.

Laparoscopy should be recommended when hysterosalpingography finds a unilateral or bilateral tubal abnormality.

Since hysterosalpingography is unreliable, laparoscopy should be recommended in cases of unexplained tubal infertility (with apparently normal hysterosalpingography).

The weaknesses of this study are:

· Interpretations of hysterosalpingography with the risks associated with inter-observer variations;

· The relatively small sample size;

· Some patients were not collected because they had hysterosalpingography, but due to lack of financial means, they were not able to benefit from a laparoscopy.

The strengths of this study are:

· Systematic performance of hysterosalpingography and laparoscopy for all patients collected;

· The multicenter nature of the study.