a Departamento de Ortopedia, Santa Casa de São Paulo, São Paulo, SP, Brasil
b Serviço de Ortopedia e Traumatologia, Hospital Israelita Albert Einstein, Universidade Federal de São Paulo (Unifesp), São Paulo, SP,Brasil
c Instituto de Ortopedia, Hospital das Clínicas, Faculdade de Medicina, Universidade de São Paulo (USP), São Paulo, SP, Brasil
d Escola Paulista de Medicina, Universidade Federal de São Paulo (Unifesp), São Paulo, SP, Brasil
Introduction
Total hip arthroplasty performed by means of a minimallyinvasive access emerged because of the trend observed inother fields of surgery. It had the aims of ensuring betterresults, fewer complications, less perioperative morbidity anda lower cost burden for public and private healthcare serviceproviders.
Several studies have shown that a minimally invasiveaccess can lead to better recovery in the immediate postop-erative period, with less pain, less bleeding during the hospitalstay, less muscle injury and, consequently, shorter hospitalstay.
However, in comparing traditional and minimally inva-sive surgical accesses, other authors have shown similarshort-term results, without any real benefit.6,7Some authorshave even warned of higher complication rates and riskswith some minimally invasive accesses, such as a singleanterior access, and have reported unusual complicationsand prolonged duration of surgery, along with long learningcurves.
Our objective here was to assess the clinical and func-tional results over the medium term (minimum follow-up ofsix years) and the complications from uncemented total hiparthroplasty performed by means of a posterior minimallyinvasive access, in comparison with a traditional right lateralaccess.
Material and methods
After obtaining approval from our institution's research ethicscommittee, a prospective comparative study was conductedin which 224 adult patients underwent elective total hiparthroplasty due to a diagnosis of primary or secondaryosteoarthrosis. The patients were selected from the waitinglist of our hospital's Hip Group.
The samples are described in Table 1.
Randomization was performed by means of a sequence ofrandom numbers, with division into two groups at the timeof making the surgical appointments. This was done by a sur-geon who was not directly involved in this study, without theknowledge of the surgeon who was the first author. In the firstgroup, the posterior access first described by Moore was used,with small-sized incisions and minimal dissection of soft tis-sues. In the patients of the control group, the traditional right
lateral access route first described by Hardinger was used, butwith the patient in lateral decubitus. There were 103 patientsin the minimally invasive access group and 121 in the rightlateral access group.
The mean length of follow-up for the whole sample was 7.2years, with a range from 6.1 to 9 years.
The criteria for non-inclusion comprised presentation ofany type of coagulopathy, systemic arterial hypertension thatwas not under pharmacological control, previous hip surgeryor sequelae from congenital hip dysplasia of Crowe types 2, 3or 4 that required structural grafting in the acetabular roof.
The exclusion criterion was loss of follow-up occurringbefore the time of the outpatient assessment in the sixth yearafter the operation.
All the patients followed the same protocol for postopera-tive analgesia, which consisted of 100 mg of tramadol (Pfizer®)intravenously every eight hours and 100 mg of ketoprofen(Eurofarma®) in a single intravenous dose daily until the day ofhospital discharge. Thromboembolic prophylaxis was admin-istered to all the patients, consisting of 40 mg of enoxaparinapplied subcutaneously (Sanofi-Aventis®), starting on the firstday after the operation and continuing until the 30th day.
All the patients were discharged on the fifth day afterthe operation and were given guidance regarding physiother-apy and anticoagulant and analgesic medications (500 mg ofparacetamol orally every six hours, if necessary).
In this study, we used two models of uncemented prosthe-sis (Meridian femoral nail with Vitalock-Stryker®acetabulumand Bicontact femoral nail with Aesculap®Plasmacup®acetabulum). The choice between these two implants wasbased merely on their availability on the day on which thepatient was operated. Both of these manufacturers were sup-pliers to our hip surgery service at the time of this study.
All the patients in the minimally invasive access groupwere operated by the first author of this study and the patientsin the right lateral access group were operated by three sur-geons, among whom was the first author.
Some points relating to the posterior minimally invasiveaccess need to be highlighted. This is an access route thatimplies a gentle learning curve, even for surgeons who areaccustomed to the traditional posterior access. The princi-ple of the moving window may be of great value with regardto acetabular and femoral access: the position of the spac-ers should be changed according to the region to be prepared,with soft-tissue displacement distally for preparation of theacetabulum and proximally for preparation of the femur, so asto avoid great tensioning of the surface tissues and minimizethe risk of scar complications at the surgical incision.
The physiotherapy protocol used consisted of assistedactive flexion of the ankles and knees using an abduction pad,starting on the first postoperative day; and gait training withpartial weight-bearing starting on the second day, with fullload applied six weeks after the operation. Each patient's evo-lution was followed up by means of return visits scheduled totake place one, three, six, 12 and 24 weeks after the operation.The return visits became annual after the first postoperativeyear.
A clinical-functional assessment was made and theradiographic criteria were also analyzed in both groups.For the clinical-functional assessment, the Harris Hip Score questionnaire was used before the operation (during the weekpreceding the surgery), six months and one year afterwardsand, finally, six years after the surgery. Six years of follow-upwas considered to be the minimum length of time for thiscriterion in the present study.
Possible late complications and aseptic loosening of theprosthetic components among the patients of this study wereevaluated after six years of follow-up.
A radiographic assessment was made at the return visit sixmonths after the operation. The acetabular positioning wasconsidered to be good if the abduction angle of the acetabu-lum was between 35?and 50?. The femoral positioning wasconsidered to be good if the femoral components presentedbetween 0?and 5?of valgus.
Presence or absence of the Trendelenburg clinical sign wasmeasured at the return visit six years after the operation.
The linear radiographic wear presented by the polyethylenewas measured using a compass and ruler marked in millime-ters at the sixth-year follow-up. The variation in thickness ofthe polyethylene between the upper and lower portions wasevaluated based on the sphericity of the prosthetic head. Weconsidered that the wear was unacceptable if it was =1 mm,which would represent more than 0.1 mm of linear wear peryear, given the length of follow-up of the patients in ourstudy.
Quantitative variables with Gaussian distribution werecompared using the non-paired Student t test. Data that didnot present normal distribution were evaluated by means ofnonparametric tests. Comparisons between the two groupswere made using the Mann-Whitney test. Qualitative (non-numerical) data were analyzed using the chi-square test, orthe Fisher test when necessary. The level of statistical signifi-cance was set at p = 0.05.
Results
Regarding the prosthesis models used, the Meridian/Vitalock(Stryker®) implant was more prevalent in both groups than theBicontact/Plasmacup (Aesculap®), but without any significantdifference (60.2% versus 53.7%; p = 0.34).
A diagnosis of primary hip osteoarthrosis was made in 61patients of the right lateral group (50.4%) and 57 of the mini-mally invasive group (55.3%), with p = 0.5. All the other patientswere diagnosed with osteoarthrosis secondary to other patho-logical conditions, which included osteonecrosis, seronegativearthritis, sequelae from Pèrthes disease, sequelae from aslipped epiphysis, rheumatoid arthritis and post-traumatichip arthritis.
The linear radiographic wear of the polyethylene was mea-sured using a compass and ruler marked in millimeters at thesixth-year follow-up. A greater degree of wear than expected(>1 mm) was observed in 30% of the patients in the minimallyinvasive group, versus 33.8% of the controls, with p = 0.56.
In the minimally invasive group, no acetabular or femoralosteolysis was identified in any of the patients from the sixthyear onwards. However, osteolysis was diagnosed in threepatients of the control group, all in relation to the acetabulum(p = 0.25).
In relation to the acetabular and femoral positioningamong the patients with a minimally invasive access, unsa-tisfactory results were noted in 5.8% of the acetabularcomponents and in 4.8% of the femoral nails. In compari-son, in the control group, there were unsatisfactory resultsin 9.9% of the acetabular components and 7.4% of the femoralcomponents, with p values of 0.32 and 0.58, respectively.
The Trendelenburg clinical test, which shows weaknessof the abductor musculature of the hip, was positive in fivepatients who were operated by means of the traditional lateralapproach but was absent from all the patients who under-went surgery by means of the posterior minimally invasiveaccess, although there was no statistically significant differ-ence (p = 0.06).
The clinical-functional assessments in the two groupsbefore the surgery were similar (mean of 47.5 in the minimallyinvasive group and 48 in the control group) (p = 0.88).
The Harris Hip Score after the operation showed that therehad been an improvement in both groups, with similar meanscores (86 in the right lateral group versus 87.6 in the mini-mally invasive group; p = 0.55).
Aseptic acetabular loosening occurred in one patient, whohad been operated by means of the minimally invasive pos-terior access. This case underwent revision 12 months afterthe initial surgery. Revision of the prosthesis was necessaryin two of the patients in the control group: the polyethylenein one patient was exchanged after 6.5 years of follow-up andacetabular revision was performed in another patient of thisgroup after seven years (p = 0.46).
The only statistically significant result was the size of theincision (p < 0.001), with a mean of 95.1 mm for the minimallyinvasive access and 169.8 mm for the traditional lateral access(Fig. 1).
Regarding complications, there were two cases in theminimally invasive group: one of medial acetabular fractur-ing without displacement, which was treated conservatively;and one of heterotopic ossification, which required surgicalresection. In the control group, there were two cases of femoralfractures, which were treated by means of cerclage during thesame surgical procedure; one case of dislocation, which wastreated by means of closed reduction, without recurrence; onecase of superficial infection, which was treated by means oforal antibiotic therapy; and one case of neuropraxia of theulnar nerve caused by the positioning on the surgical table,which regressed after eight weeks. No significant differencewas observed between the groups (p = 0.45).
Discussion
In our study, we noted that the mean age of our patients waslower than seen in studies in the literature, which mostlyshowed mean ages greater than 60 years.2,11This can per-haps be explained by the considerable numbers of cases ofosteoarthrosis secondary to systemic and rheumatologicaldiseases that were observed in our setting (almost half of ourpatients). Another possible cause for this finding was perhapsthe typical pyramidal age distribution of the Brazilian popula-tion, which differs from that of European or North Americancountries. This lower mean age may represent a source of
bias, both toward better recovery from the acute postopera-tive phase and toward the complications inherent to wear andloosening of the arthroplasty, given that these patients theo-retically have a higher degree of physical and work activity.
Other possible sources of bias in the results obtained in thepresent study include the fact that two prosthesis models wereused in both groups, although both models were uncemented.However, most of the studies with large samples have tendedto present this same bias. Regarding the control group, thefact that the procedures were performed by different surgeonsmight also be considered to be a source of bias, although diver-sity of surgeons has also been reported by a good proportionof previous authors.
The use of prostheses with a tribological pair consisting ofa conventional polyethylene insert and a metal head, togetherwith the patients' lower age and consequently higher level ofactivity, may have been responsible for the linear wear greaterthan expected that was found in one-third of the cases, andfor the early acetabular osteolysis that was found in threepatients of the control group. However, we emphasize that themethodology used for measuring the wear of the polyethylenepresents a margin of error of around 15%. Methods of greaterprecision such as computerized three-dimensional models areused today, but these were not taken into consideration in thepresent study because they did not form the authors' mainobjective.
Regarding the positioning of the prosthesis components,although this was not the main objective of our study, we didnot observe any statistically significant difference between thegroups. However, in our previous experience, larger acetabula
(>52) tend to present greater technical difficulty for surgeons,with a higher risk of lateralization and greater abduction anglefor the acetabular component. This results from the angle ofattack of the acetabular reamer, which is harmful particularlywhen reamers and impactors that are adapted for smalleraccesses are not used (Fig. 2).
Weakness of the abductor musculature, as represented by apositive Trendelenburg test, was only found in patients in thegroup operated by means of a lateral access. This result wasnot significant (p = 0.06). Kinematic gait studies have proventhat better results are obtained when the access used doesnot interfere directly with the integrity of this musculature.
Most studies have shown that the acute postoperativerecovery is better when a minimally invasive access is used,with less bleeding, lower muscle damage and shorter hospi-tal stay. The question of bleeding during the hospital stay isanother polemical point because although most authors havereported mean volumes of bleeding of 500 mL,2,3,14,15the inci-dence of transfusions among the patients is close to 50%. Thisdenotes a contradiction and a likely underestimate, whichhas been confirmed by other studies in which the centralscope was blood volume loss and the methodology was moredetailed for such estimates.
Our study, with a follow-up of more than six years,presented results similar to data in the literature on thistopic, as demonstrated in a recent meta-analysis, whichshowed similar evolution in the two groups, both in rela-tion to the radiographic parameters and in relation to theclinical-functional score of the Harris Hip Score, at all thetimes evaluated. However, the medium and long-term follow-up seems to be uncertain and inconclusive, given the scarcityof comparative studies with longer follow-ups, and there is noevidence so far regarding the superiority of these accesses inrelation to the traditional routes.
Overall, in the whole sample, we only observed one case ofdislocation. We believe that this low incidence was due to theintensive work conducted by the physiotherapy team of ourservice, in which postural educational measures were imple-mented to diminish the risk of this complication. Moreover, in our study, we were only dealing with patients with a diagnosisof osteoarthrosis. It is well known that these patients presentlower risk than do patients with a diagnosis of fracturing ofthe femoral neck who undergo total hip arthroplasty.
We chose to evaluate overall occurrences of complicationsper group, in a comparative manner, because the sample sizewould be insufficient for us to evaluate each comparison indetail according to each type of complication that occurred.
The only variable evaluated in this study that showed asignificant difference for the minimally invasive access, i.e.the smaller-sized incision, theoretically would indicate thatthe esthetic result would be better. However, in 2005, Mowet al.19demonstrated worse scarring results from small inci-sions. In our sample, there were no esthetic complaints, butit needs to be borne in mind that this should be last criterionfor indicating a less invasive access.
Conclusion
The medium-term clinical and radiographic results and thecomplication rate are similar for patients undergoing totalhip arthroplasty by means of a posterior minimally invasiveaccess and by means of the traditional right lateral route.However, the right lateral access more frequently causesinsufficiency of the abductor musculature, in relation to theposterior minimally invasive access.
Conflicts of interest
The authors declare no conflicts of interest.
1. Wenz JF, Gurkan I, Jibodh SR. Mini-incision total hiparthroplasty: a comparative assessment of perioperativeoutcomes. Orthopedics. 2002;25(10):1031-43.2. Goldstein WM, Branson JJ, Berland KA, Gordon AC.Minimal-incision total hip arthroplasty. J Bone Joint Surg Am.2003;85 Suppl. 4:33-8.3. Matta JM, Shahrdar C, Ferguson T. Single-incision anteriorapproach for total hip arthroplasty on an orthopaedic table.Clin Orthop Relat Res. 2005;441:115-24.4. Berger RA. Total hip arthroplasty using the minimallyinvasive two-incision approach. Clin Orthop Relat Res.2003;(417):232-41.5. Vicente JR, Croci AT, Camargo OP. Blood loss in the minimallyinvasive posterior approach to total hip arthroplasty: acomparative study. Clinics. 2008;63(3):351-6 [Sao Paulo].6. De Beer J, Petruccelli D, Zalzal P, Winemaker MJ.Single-incision, minimally invasive total hip arthroplasty:length doesn't matter. J Arthroplasty. 2004;19(8):945-50.7. Ogonda L, Wilson R, Archbold P, Lawlor M, Humphreys P,O'Brien S, et al. Minimal-incision technique in total hiparthroplasty does not improve early postoperative outcomes.A prospective, randomized, controlled trial. J Bone Joint SurgAm. 2005;87(4):701-10.8. Archibeck MJ, White RE Jr. Learning curve for the two-incisiontotal hip replacement. Clin Orthop Relat Res. 2004;(429):232-8.9. Bal BS, Haltom D, Aleto T, Barret M. Early complications ofprimary total hip replacement performed with a two-incisionminimally invasive technique. J Bone Joint Surg Am.2005;87(11):2432-8.10. Griffith MJ, Seidenstein MK, Williams D, Charnley J. Socketwear in low friction arthroplasty of the hip. Clin Orthop RelatRes. 1978;(137):37-47.11. Swanson TV. Early results of 1000 consecutive, posterior,single-incision minimally invasive surgery total hiparthroplasties. J Arthroplasty. 2005;20(Suppl 3):26-32.12. Vicente JR, Croci AT, Camargo OP. Restaurac¸ão do centro derotac¸ão na artroplastia total do quadril minimamenteinvasiva. Acta Ortop Bras. 2009;17(2):14-7.13. Ritter MA, Harty LD, Keating ME, Faris PM, Meding JB. Aclinical comparison of the anterolateral and posterolateralapproaches to the hip. Clin Orthop Relat Res. 2001;(385):95-9.14. Chimento GF, Pavone V, Sharrock N, Kahn B, Cahill J, SculcoTP. Minimally invasive total hip arthroplasty. A prospectiverandomized study. J Arthroplasty. 2005;20(2):139-44.15. Wright JM, Crockett HC, Delgado S, Lyman S, Madsen M,Sculco TP. Mini-incision for total hip arthroplasty. Aprospective, controlled investigation with 5-year follow-upevaluation. J Arthroplasty. 2004;19(5):538-45.16. Rosencher N, Kerkkamp HEM, Macheras G, Munuera LM,Menichella G, Barton DM, et al. Orthopedic surgerytransfusion hemoglobin European overview (OSTHEO) study:blood management in elective knee and hip arthroplasty inEurope. Transfusion. 2003;43(4):459-69 [Paris].17. Pierson JL, Hannon TJ, Earles DR. A blood-conservationalgorithm to reduce blood transfusion after total hip andknee arthroplasty. J Bone Joint Surg Am. 2004;86(7):1512-8.18. Moskal JT, Capps SG. Is limited incision better than standardtotal hip arthroplasty? A meta-analysis. Clin Orthop RelatRes. 2013;471(4):1283-94.19. Mow CS, Woolson ST, Ngarmukos S, Park EH, Lorenz PH.Comparison of scars from total hip replacements done with astandard or a mini-incision. Clin Orthop Relat Res.2005;(441):80-5.
