Total hip arthroplasty (THA) has transformed orthopaedic surgery, enabling the restoration of joint architecture and biomechanics while preserving the integrity of the periarticular musculoskeletal structures. While it provides excellent long-term functional outcomes [1], the long-term survival of the implant is dependent on the absence of complications [2].

Surgical revision is therefore imperative in order to restore joint function and correct mechanical or biological failures. Currently, there is a constant increase in revision rates, driven by rising life expectancy and the implantation of prostheses in younger, more active patients [3]. The causes of these revisions are primarily dominated by aseptic loosening, bearing surface wear, instability and periprosthetic infection [4]. In West Africa, studies from Ghana [5] and systematic reviews across sub-Saharan Africa [6] have reported revision rates of 4% - 8%, with aseptic loosening, dislocation, and periprosthetic fractures as leading causes.

Against this backdrop, our study aimed to identify and analyze the causes of revision for primary THA in a single-center series at Idrissa Pouye General Hospital in Senegal, while discussing their relevance to the broader sub-Saharan African context.

- Study setting and design

We conducted a retrospective, descriptive and continuous study over an 11-year period (1 January 2009 to 31 December 2019) in the Orthopaedics and Traumatology Department at Idrissa Pouye General Hospital in Dakar.

- Data collection and inclusion criteria

Data were collected from operating room registries, hospitalisation records and outpatient consultation files. All patients who underwent a THA revision were included, regardless of gender. Patients with incomplete records (missing clinical notes or radiographic imaging documenting the complication) were excluded.

- Population characteristics

During the study period, 552 total hip replacements were performed in the department, and revision procedures accounted for only 5.43% of all total hip replacements performed. The revision rate for primary THAs implanted in our department during the study period was 3.40%.

The series included 30 revisions two of which were bilateral (28 patients). The mean age at the time of surgery was 56 ± 15 years. There was a male predominance of 60.70%, with a sex ratio of 1.5. The left side was involved in 60% of cases. Most prostheses were implanted at Idrissa Pouye General Hospital (63.3%). Other hospitals in Senegal treated nearly a quarter of the cases (23.4%). A significant proportion (13.3%) were implanted abroad.

Baseline data regarding the revised hips (primary indication, type of fixation, surgical approach, and implant characteristics) could not be systematically collected, since a proportion of the patients had undergone their primary surgery in other institutions.

- Study parameters and classifications

The study focused specifically on the aetiologies of revision. For lesion analysis, the following reference tools were used (Table 1):

loosening: classified according to the SOFCOT criteria and the Paprosky classification for assessing bone stock;periprosthetic fractures: Evaluated according to the Vancouver classification.

Table 1. Diagnostic criteria for THA complications.

- Overall etiological analysis:

Aseptic loosening was the leading cause of revision in our series (46.67%), followed by prosthetic dislocation (26.67%). Periprosthetic fractures accounted for 13.3% of cases (see Table 2). Other causes included material wear or failure (6.67%), fracture-dislocation (3.33%), and septic loosening (3.33%).

- Loosening

Loosening was aseptic in 46.67% of cases and septic in 3.33%. The mean time to onset for aseptic loosening was 10 ± 8 years.

Loosening was predominantly bipolar, accounting for 80% of cases (Figure 1). It involved the acetabular cup in 51.85% of cases and the femoral stem in 48.15% of cases.

According to the SOFCOT classification (Table 3), Type 2 was the most frequent in both the femur (33.33%) and the acetabulum (20%). Regarding bone loss, Paprosky grades 2A and 2B were the most prevalent on both sides (Table 4).

Table 2. Causes of revision.

Table 3. SOFCOT classification of loosening.

Table 4. Paprosky classification of loosening

- Dislocations

Prosthetic instability was observed in eight cases (26.67%), including one bilateral case (Figure 1). Dislocation was the second most common cause in this study. All dislocations were classified as early (< 5 years), with a mean time to occurrence of 2.93 months (range: 2 days-1 year).

- Periprosthetic Fractures

Four patients (13.33%) underwent revision for a fracture involving a THA. According to the Vancouver classification, all of these fractures (100%) were Type B2, involving a bony break associated with an unstable femoral implant (Figure 2).

- Wear and Mechanical Incidents

Two cases of material wear and degradation were observed: an “explosion” of the polyethylene insert following trauma (Figure 3), and breakage of an acetabular

Figure 1.Septic prosthesis loosening with cup migration into the iliac wing.

Figure 2.Periprothetic fractures Vancouver type B2.

Figure 3. Shattered prosthesis liner (arrow).

fixation screw.

Aseptic loosening was the primary cause of revision in our series (46.67%). This predominance is widely documented in the international literature. In the United States, Bozic et al. [7] reported a rate of 51.70%, while French series such as those by Gasteau et al. [8] (63.20%) and Lebeau et al. [9] (65.90%) confirm this trend. Our results are also similar to those of Oltean-Dan et al. [10], who, in their study of a regional orthopedic center, identified aseptic loosening as the major cause of revision (44.18%). This complication occurs preferentially in young, highly active subjects due to mechanical stress, or in elderly subjects whose osteoporosis weakens the bone-prosthesis interface.

The mean time to onset in our study was 10 ± 8 years. This result is consistent with data from Fen et al. [11] (10 years) and falls within the average observed by Oltean-Dan et al. [10] (8.4 years). The longevity of a THA remains closely linked to the precision of the surgical technique, the quality of the bone stock, and patient compliance.

The relatively young profile of our population (mean age of 56 years) leads us to compare our results with the work of Kahlenberg et al. [12], who emphasize that in active patients, instability and early wear are predominant causes of revision. Similar to the Idrissa Pouye General Hospital, the study by Oltean-Dan et al. [10] conducted in a regional center shows that these referral facilities attract a high volume of severe mechanical pathologies, where the management of primary prosthesis failure becomes a major surgical activity.

Dislocation accounted for 26.67% of our failures. This rate is higher than the 22.5% reported by Bozic et al. [7] and the 13.9% observed by Oltean-Dan et al. [10] In our context, this instability may be linked to anatomical factors (abductor insufficiency), technical factors (implant malposition), or non-compliance with postoperative precautions by young patients eager to quickly return to normal activities.

A notable feature of our series is that 100% of the periprosthetic fractures (13.33% of cases) were Vancouver Type B2. This figure is significantly higher than the 5.80% fracture rate reported by Bozic et al. [7]. The systematic occurrence of Type B2 (unstable implant) suggests either high-energy trauma or the existence of chronic, unrecognized loosening that weakened the femur.

This severity is also reflected in the analysis of loosening, where SOFCOT Type 2 and Paprosky Grade 2A predominated, matching the observations of Lebeau et al. [9]. These significant bone losses are often the result of delayed diagnosis. As noted by Oltean-Dan et al. [10], prolonged consultation delays in regional centers explain why bone destruction is often advanced at the time of management, making revision surgery technically complex.

Finally, our septic loosening rate (3.33%) is remarkably low compared to the 14.80% reported by Bozic et al. [7] and the 23.20% reported by Oltean-Dan et al. [10]. In the sub-Saharan African context, where infection control remains a challenge, this finding highlights the importance of rigorous perioperative protocols and suggests that regional centers may achieve outcomes approaching international standards. However, given the retrospective design and small sample size, the low rate of septic loosening should be interpreted descriptively rather than as definitive evidence of aseptic mastery.

This 11-year study confirms that aseptic loosening is the primary cause of THA revision at the Idrissa Pouye General Hospital (46.67%), with an average implant survival of 10 years.

However, two major challenges stand out: the high frequency of dislocations (26.67%), linked to the young and active profile of the patients, and the severity of periprosthetic fractures (100% Vancouver Type B2), underscoring the predominance of unstable implants and the technical challenges they pose during revision surgery. Conversely, the low rate of infectious complications (3.33%) highlights the quality of the initial management.

Improving outcomes will require increased precision during primary surgery and the establishment of rigorous radioclinical follow-up, which is essential for detecting mechanical failures before the onset of massive bone loss.