The final results on LLD of our patients post THA on the post-OP x-ray are comparable to the results of other studies using a CAL intraoperatively [10, 12]. There also is a good correlation of the intraoperative measurements of the leg length and total offset using the calliper in reference to the values measured on the antero-posterior x-ray of the pelvis. As shown in the literature, the use of a calliper intraoperatively significantly improves the results on leg length and offset in comparison to just using clinical tests or templating [10, 11, 12]. Similarly with all other studies, we also considered the following targets: no leg length discrepancy and the restoration of the offset on the x-ray pelvis taken postoperatively. As mentioned, additional clinical tests to check stability and impingement were done intraoperatively. There were no patients where it was necessary to change the neck length or other prosthesis parts to prioritize stability over achieving the scheduled limb-length and offset, abandoning the calliper input. In agreement with Barbier et al., the calliper seems to better predict the leg length than the offset [12]. Our study also confirms that good results using the calliper can be achieved with other surgical approaches as we used a different approach than Barbier et al. or Enke et al. [10, 12].
In his two keystone publications on LLD, Gurney et al. concluded that it is generally challenging to find a perceptive, functional and anatomic patient treatment benchmark for LLD [3, 13]. Options to treat LLD in adults include physiotherapy guided exercises, modifiable heel lift or post-surgical interventions [3]. Although there might be a break point around a LLD of 20 mm, every patient presenting a LLD needs to be considered individually on a case-by-case basis. Patient communication and education should include discussing the different treatment options. There was no incidence of return to surgery for revision with any patients included within this study related to postoperative LLD.
Among our patients were several with severe pelvis pathology or posttraumatic avascular caput necrosis presenting preoperatively as an LLD of around − 2.9 cm, for the leg to be operated shorter than the contralateral side. Excessive limb lengthening increases the odds for nerve palsy [14]. Therefore, our target of reaching complete correction of LLD was compromised. In these cases, even experienced hip surgeons find it difficult to evaluate the amount of increase of LL intraoperatively [15]. We found the calliper to be especially useful during operations when facing these challenges.
There are different CALs on the market, using the same principle with a fixed reference point on the pelvis and a mobile point on the femur [6, 10]. They all are similar and function successfully. This study utilized the CAL from Smith & Nephew because of our long-lasting experience with it. It is not cost-prohibitive in relation to instrument supply and has reliable international availability. The literature and the results of our study show convincingly that the use of an intraoperative calibration gauge improves significantly the postoperative results concerning LLD [5, 10, 12].
Within the literature other authors have expressed doubts that this is also the case concerning the offset [12]. One apparent reason is the fact that the calliper measurements are related to a change in “total offset” and not in “femoral offset”, which seems to be more important for the functional outcome after THA [4, 16]. We compared the postoperative total offset with the preoperative offset of the same hip. In the literature there is disagreement about which value should be used as reference when comparing postoperative offset after THA. Some authors use as reference the preoperative offset of the operated hip [4, 17], some use the offset of the contralateral not operated hip [16], whilst others use absolute values as 42–48 mm [18].
Similar to some studies, we chose to measure the change of offset on digital radiographs of the pelvis [10, 12, 17]. We consider this a major weakness of this study. Reproducibility is poor, with a mean error of about 9.7 mm, which means 22% assuming an offset of 45 mm, when using x-ray for measuring offset [20]. While CT seems to be the golden standard to measure offset [19], already a software upgrade of the digital x-ray gives more precise results Einbildröntgenanalyse [4]. Due to software incompatibility with our existing PACS system this upgrade was not used for this study.
Although interest in postoperative offset after THA is increasing, it is still unconfirmed that differences in offset lead to severe complications. In a meta-analysis study, De Fine et al. did not report any difference in bearing surface wear, implant loosening or dislocation rate when looking at THA patients with different femoral offsets using hard bearing surfaces [21]. A study from the New Zealand registry does not show any difference in the revision rate related to different offsets due to different designs of uncemented femoral stems [18]. Studies investigating clinical results after THA report the growing tendency that patients with decreased femoral offset are more Trendelenburg positive and have a worse outcome on the Oxford Hip Score (OHS) than patients with a restored or increased offset [4, 9, 22, 23]. This might emphasize the importance of restoring the abductor lever arm to at least a certain length [16]. Summarizing these results, in case of indecision rather an increase in offset might be recommended [4, 9, 16].
In our study, we saw a strong correlation between the results of our intraoperatively used calibration gauge with the measurements on the x-ray pelvis concerning change of offset r = 0.542, p < 0.01. Despite reviewing available literature, we still have challenges with a proper interpretation of these results in relation to the benefit for the patient. We cannot even define a clear target zone for postoperative offset. In the clinical setup, any LLD seems to be more relevant than any discrepancy in offset. Therefore, when intraoperatively a situation arises where a mismatch in LL and offset has to be compromised, the LL should be made accurate. Further studies will be required to clarify this situation.
Postoperative migration and subsidence of both components, the acetabular and the femoral component, are known and well-studied phenomenona after THA surgery. The Corail stem, used for 57 of our 59 operated hips, is uncemented and fully hydroxyapatite coated. According to Selvaratnam et al., most of the subsidence of this stem occurs within the first 6 weeks after the operation [24]. In literature the amount of postoperative subsidence of the stem and the percentage of patients concerned with it, greatly differ. Faisal et al. report that the uncemented collarless Corail stem can be used safely for all patient groups, even the elderly, and almost no subsidence occurs [25]. Using the same stem, Ries et al. found a mean subsidence of 3.1 mm after a mean follow-up of 7 months [26]. Other authors report that about 30% of the stems subside more than 3 mm within the first 6 weeks [24, 27]. It might be considered that the use of a collared stem is the answer to avoid stem subsidence. However, in clinical series, only about 39% of collars have primary contact to the bone [28]. In a large series, it is shown that the mean subsidence of uncemented stems is 2.9 mm (0-20.4 mm) whereas the addition of a collar leads to a lesser degree of subsidence but does not avoid it [29]. A collarless uncemented stem respects the press-fit principle and, therefore, should lead to better bony osseointegration of the prosthesis and less aseptic loosening. It should be mentioned that even a cemented femoral stem, the Exeter stem, has a subsidence of 1.42 mm (0.43–3.91) within the first two years and afterwards 0.08 mm/year [30].
The Pinnacle cup, used for 57 of our 59 operated hips, is uncemented, hemispheric and fully hydroxyapatite coated. According to Dammerer et al., within the first 2 years this cup shows a mean total migration of 1.42 mm (0.1–6.3) [31]. Another study using a similar cup shows an even higher migration within the first two years postoperatively for patients being operated for rheumatoid arthritis 2.62 mm (0.55–8.22) than for osteoarthritis 1.44 mm (0.1–5.62) [32]. Nieuwenhuijse et al. report that even the cemented Exeter cup shows a mean cranial migration of 0.94 mm within the first 2 years [33].
Until now all studies evaluating leg length after THA are based on the target that, directly postoperatively, the patient should have the same leg length as the contralateral side [34]. The patients´ expectations are set up the same way. Even though the leg length postoperatively might be precisely the same, the patient often has a feeling immediately after the operation that the operated extremity is “too long”. This may be due to postoperative swelling and pain and usually disappears within the first weeks. Another study reveals that even at the long view 64% of patients after THA perceive a LLD despite radiologically this is not apparent, setting the threshold for LLD at \(\ge\) 5 mm [1]. The list of states that can lead to a perception of LLD is long: spine pathology, pelvic obliquity, LLD concerning status post THA, neurological impairment, knee malalignment [35] and other discrepancies below the hip, leading to a very complex situation. According to the authors, there is an additional time dependent factor to be considered: the migration of the prosthesis components, which contributes to a change in perception of LLD. Summarizing the studies named above, patients can easily expect a change in LLD of 5 mm or more sometime after surgery due to subsidence of the stem and the cup. Subsidence of prosthesis components after THA is common. A certain amount of subsidence and long-term shortening of the extremity should not be evaluated as a complication. It should be seen as the prosthesis's natural process when embedded in a living environment. In our opinion, the orthopedic surgeon, the patients, and legal institutions have to consider these factors when evaluating the success of a THA operation. This must be reflected upon intensively in pre- and postoperative counselling and be part of the surgical consent of the THA patient.