If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Pathological dislocation of hip following episode of septic arthritis is a long recognized complication. We determined clinicoradiological outcome in 14 children presenting with acute septic dislocation of hip and reduced at the time of emergency arthrotomy and drainage.
Methods
The retrospective study was conducted at a tertiary care health centre located at suburb of a low income country. The study involved a 10 years chart review of children with confirmed septic arthritis of hip presenting with complication of acute joint dislocation and treated with open reduction at the time of arthrotomy and surgical drainage. All included patients were provided postoperative abduction splintage for minimum 3 months. The clinical evaluation at final follow up was done using modified Moon’s criteria for outcome assessment. The radiological outcome was analyzed in terms of containment (Severin class), sequelae (Choi radiological types) and effect on acetabular development (acetabular index, AI). The AI on normal and reduced sides was compared statistically.
Results
Average patient age at the time of presentation was 7.2 months and follow up available was 32 months (range, 12–70 months). Functionally, 5 had excellent and 6 had good outcomes at the final follow up. Two patients had fair whereas 1 had poor outcome. Seven hips were Severin class II, 4 class III and 2 class IV and 1 was class VI. One patient had Choi type IA, 3 type IB, 8 type IIA, 1 type IIB and only 1 had type IVA at follow up. The mean AI of reduced hip was 23.4° and was significantly higher than that of unaffected side (mean 15.7°).
Conclusions
The complication of acute septic dislocation of hip in children may not have a sinister prognosis as commonly believed. These hips however were associated with significant acetabular dysplasia in the follow up period.
Septic arthritis of hip is a suppurative infection, most often of haematogenous origin. The purulent joint effusion associated with the condition may result in pathological subluxation or dislocation, if the increased intraarticular pressure persists for several days.
Despite being a known complication of septic arthritis hip for a long time, there are only few studies in literature which document outcome in acute septic dislocation.
We questioned whether the sinister tag associated with this finding still holds true in this era of modern medicine when emergency arthrotomy and prompt broad spectrum antibiotics administration is a norm. To find an answer, we retrospectively studied clinicoradiological outcome in 14 patients with septic dislocation of hip presenting to our institution over a span of 10 years. The radiological outcome was analyzed in terms of containment, aftermaths of infections (sequelae) and effect on acetabular development.
2. Methods
The retrospective study was conducted at a tertiary care health centre located at suburb of a low income country. The study involved chart review (January 2010–December 2019) of children with confirmed septic arthritis of hip presenting with complication of acute joint dislocation and treated with open reduction at the time of arthrotomy and surgical drainage (Table 1). We included those cases where the pathological dislocation was documented using ultrasonography and/or radiographs [International Hip dysplasia Institute (IHDI) Classification] and intraoperative findings and a clinicoradiological follow up beyond 12 months was available. All included patients underwent emergent arthrotomy within 48 h after presentation to our institute. Fourteen children fulfilled the inclusion criteria and were included.
The arthrotomy was performed using Watson Jones anterolateral approach. Femoral head was reduced in the acetabulum after through joint lavage. Negative suction drains were left for 24–48 h. Hips were protected in reduced position for 3–4 weeks by application of hip spica cast or T-slab (in case of neonates) followed by abduction bracing for at least 3 months till healing was advanced. All patients were initiated on broad spectrum antibiotics which were later changed as per culture report. The minimum duration of antibiotics prescribed was 6 weeks.
4. Evaluation at follow up
The clinical evaluation at final follow up was done using modified Moon’s criteria for outcome assessment in view of squatting and cross legged sitting practices of the region.
An ‘excellent’ outcome was defined as pain free hip and normal ambulation, sitting possible in any position; ‘good’ as slight pain, occasional, no compromise in activities, uneasy squatting; ‘fair’ as mild pain, no effect on average activities, rarely moderate pain with unusual activity, some limitation in cross legged and squatting and ‘poor’ as moderate and marked pain, limitation of ordinary activity and serious limitation of activities. For comparison of results, patients with Moon’s excellent or good results were grouped as satisfactory, and patients with fair or poor results were grouped as unsatisfactory.
Serial radiographic evaluation was based on plain X-rays. The overall containment was assessed by modified Severin classification and hips were classified according to Choi’s radiological types for the residual sequelae.
The Choi 1, 2, and 3A types were considered satisfactory (femoropelvic articulation present); 3B and 4 unsatisfactory for the purpose of this study as these types had femoropelvic dissociation and always demanded major surgical reconstructive measure.
Acetabular index (AI) was calculated for both sides in the follow up radiographs to document the acetabular dysplasia.
5. Statistical analysis
The statistical analysis was done on the online ‘MedCalc’ statistical software. The AI on normal and reduced sides was compared using unpaired student t-test. P values of <0.05 were considered significant.
6. Results
Average patient age at the time of presentation was 7.2 months (range, 0.5–36 months) and follow up available was 32 months (range, 12–70 months). Out of 14 patients, 12 were infants at initial presentation. Two out of 14 had septic affection of both hips but the contralateral side has not been included in calculations as it was not initially dislocated.
Infection was controlled in all hips. Varus femoral osteotomy was performed in 1 patient at the age of 50 months to attain better containment. Functionally, 5 had excellent and 6 had good outcomes. Two patients had fair whereas 1 had poor outcome as per Moon’s criteria at final follow up. As per Severin’s classification, 6 hips had only mild deformity of femoral head, neck or acetabulum (class II), 4 were class III (dysplasia), 3 were class IV (subluxation) and only 1 had a redislocation (class VI). According to Choi classification, 1 patient had type IA, 3 had type IB, 8 had type IIA, 1 type IIB (Fig. 1) and 1 had type IVA at follow up.
Fig. 1A) 2 month old child with septic dislocation of left hip (patient 10) (International Hip Dysplasia Institute (IHDI) classification grade III). B) 6 months follow up. C) 36 months follow up, capital femoral epiphysis still small and irregular. D) Radiographs at final follow up at 62 months. Femoral head showing Choi type IIB changes. The Severin class was II. The acetabular dysplasia was manifest by an AI of 27° on the affected side compared to 17° on normal side. The horizontal/vertical/oblique dashed lines respectively represent the H line (Hilgenreiner’s line), P line (Perkin’s line) and D line (diagonal line). The white dot represents the H point which is defined as the midpoint of the superior margin of proximal femoral metaphysis.
The mean AI in 14 hips (dislocated at initial presentation) on the follow up radiographs was 23.4 ± 5.7° (range, 14–35°) and on the unaffected side in 12 unilateral cases was 15.7 ± 4.0° (range, 11–23°). The difference between these two was statistically significant (p = 0.0007). The AI on contralateral side in patients with bilateral hips was 17° (patient 5) and 29° (patient 8).
7. Discussion
7.1 Principal findings
Our series had 12 infants, but the complication of acute dislocation was seen in older children as well. The complication therefore occurred irrespective of age. All hips were managed by uniform protocol of emergency arthrotomy, drainage and gentle open reduction. At follow up, the clinical outcome was satisfactory in 11 out of 14 patients (79%). The reduction obtained at arthrotomy was maintained in all but one which eventually had a frank re-dislocation. This hip had a class VI Severin hip (re-dislocation) and 3 others had class IV (subluxation). All other 10 hips (72%) had a Severin class I-III. The dislocated hips (after reduction) changed mostly to Choi type I or II, again indicating a largely satisfactory outcome (Fig. 1). There was only 1 case of Choi IVA sequelae in our series. One key finding was presence of significant acetabular dysplasia on reduced side at follow up. Obviously, the hips were not dislocated for prolonged periods to induce this dysplasia. Therefore, the dysplasia seemed a combined effect of changes in proximal femur and acetabular roof due to infection. To explain further, the septic arthritis of the hip is a panarticular phenomenon with affection of both joint surfaces. The affection of acetabular side possibly manifests itself as dysplasia. In two patients with bilateral hip affection, the acetabular dysplasia on contralateral side matched the dysplasia on reduced side supporting the above proposition.
7.2 Comparison with existing literature
The acute dislocation as a complication of septic arthritis in infants is a frequent finding. In a series of 14 septic joints affected under 3 months of age, Bono et al. reported hip subluxation/dislocation in 2 out of 3 hips.
Since previous century, the presence of septic dislocation of hip is recognized as a poor prognostic sign. Badgley et al. (1936) described 113 cases of septic hip in the pre-antibiotic and non-emergency arthrotomy era.
The dislocation was present in 34/113 cases in his series. The authors concluded that dislocations were less frequent if the treatment was instituted early and limb was kept in extension and abduction. Among end result, 6 patients died and 11 had persistent dislocation. Only 17/34 cases had a reduced hip in long term. Out of these reduced cases, good function was reported only in three resulting in the conclusion that this complication of septic hip resulted in a poorly functioning joint. Ngom et al. studied 7 recent (<3 weeks) and 12 late (>3 weeks) septic hip dislocations.
They advised arthrotomy in recent dislocations and obtained good results in all such cases. All these hips escaped major lesions/deformations of femoral head (coxa brevia/valgum/magna/plana, osteochondritis, loss of femoral head etc.). The authors advised 5 weeks of traction after arthrotomy followed by hip spica cast for 8 weeks to maintain reduction. In another study from Japan, 9 children with septic arthritis of hip were treated with emergent arthrotomy and drainage.
There were 4 infants in this series and all exhibited hip dislocation at diagnosis. The end result at mean follow up of 5.6 years, the dislocated hips resulted in Choi type IIA hips in 2 patients and IIIA and IVA in one patient each.
Our postoperative protocol of a shorter duration of plaster cast followed by less rigid immobilization in brace was successful in keeping hip reduced as well as providing function.
We had excellent to good functional results in 79% patients. The finding of less severe hip sequelae following emergency management in our study is in line with reported recent literature.
The outcome of acute septic hip dislocations has thus improved much in this era of modern medicine.
7.3 Clinical implications
Once joint is decompressed, the hip tends to reduce and stabilize. Most of these hips therefore can be managed with usual protocol of emergent drainage and antibiotics. The only care required is postoperative splintage in abduction to aid containment. The late effects of acute dislocated hips in sepsis are apparently because of changes occurring in proximal femur and acetabulum rather than dislocation itself. However, the concomitant significant acetabular dysplasia noted in these hips does indicate a pan-articular joint affection. Longer follow up is needed to establish the eventual outcome of acetabular dysplasia due to septic dislocation.
7.4 Research implications
It is possible that the acetabular dysplasia noted in these hips may resolve or deteriorate with child’s growth. Therefore these hips demand a longer supervised follow up.
7.5 Limitations
We are well aware that a sample size of 14 patients is relatively small which may not be true representative of disease profile in general population. Further, the results of the study should be interpreted in context of it being retrospective in nature. We did not include the initial duration of symptoms to arthrotomy, which were considered imprecise being largely based on parent’s/guardian’s recall and their understanding of symptoms. Also patient’s prior treatment history before reaching our institution was not documented as it was ineffective in controlling infection (collections seen in all patients at the time of drainage). The organism’s virulence and whether it was hospital or community acquired was not investigated in the study. The head deformation/destruction post sepsis was accounted for by Choi’s types since it may be because of several reasons including avascular necrosis, destruction by bacterial toxins, involvement of growth plates, etc. It is often difficult to determine the precise reason for the same.
7.6 Strengths
The strengths of our study were it being a focussed study detailing outcome in acute septic dislocation hip in children. The study had a reasonable follow-up of average 32 months at which most of the sequelae of suppurative pathology have already been manifested.
This is probably the only study in children which quantitatively determined impact of infection on proximal femur and acetabulum in these hips.
7.7 Conclusions
The complication of acute septic dislocation of hip in children may not have a sinister prognosis as commonly believed. These hips however were associated with significant acetabular dysplasia in the follow up period.
Financial conflicts
Nil.
Not presented anywhere.
Declaration of competing interest
Nil.
References
Hasegawa Y.
Ito H.
Intracapsular pressure in hip synovitis in children.
30570-1&id=gr1.jpg){kind=link}