Introduction: Studies about the influence of patient characteristics on mechanical socket failure in THA apply different methodologies and reveal inconclusive results. In addition the fixation mode has rarely been in the focus of investigation.

Methods: We conducted a two-arm matched pairs case-control study assessing the influence of patient characteristics sex, age, weight, BMI, diagnosis, and activity level (Charnley classification) on the odds for mechanical socket loosening. The cemented and uncemented fixation mode was analyzed separately.

Results: 299 cases and 986 controls were included in the cemented study arm and 510 cases and 3000 controls in the uncemented arm. Women had reduced loosening odds for the cemented (OR=0.59, p=0.0024) and uncemented (OR=0.63, p=0.0001) fixation compared to men (OR=1). Each additional year the intervention was postponed reduced the loosening odds by about 2 % for cemented (OR=0.98, p=0.017) and uncemented (OR=0.98, p=0.0002) sockets. In cemented sockets the weight group of 73–82 kg had lower loosening odds (OR=0.63, p=0.017) than the lighter (OR> 0.92) and heavier (OR> 1.1) weight groups. There were no significant effects of weight in the uncemented group. In contrast, obese patients (BMI> 30) with uncemented sockets displayed elevated loosening odds (OR=1.41, p=0.034) compared to an insignificant effect in the cemented arm. Osteonecrosis was the only main pathology revealing elevated loosening odds (OR=1.27, p=0.049) but only in cemented sockets. The Charnley classification as an indirect proxy of activity revealed changing and insignificant effects.

Conclusions: Female sex and a delayed intervention have similarly protective effects on the odds for cemented and uncemented socket loosening. Whilst a certain body weight range has a significantly protective effect in cemented sockets the more important finding is the significantly increased risk for uncemented socket loosening in obese patients. Patients with osteonecrosis are the only etiologic group at a significantly increase risk for socket loosening, but only with a cemented fixation. The Charnley classification as a surrogate for activity level had no influence on loosening risks in either socket fixation mode. Level of Evidence: Level III (case-control study)

Experimental work has shown that dislocation of the shoulder may involve disruption of the capsule from its lateral humeral attachment. We report two patients with recurrent dislocation due to this injury. Lateral repair gave good results. It is suggested that this injury be considered and looked for when glenoid labral injury is minimal or absent.