Comparison of Direct Anterior Approach and Posterolateral Approach in Total Hip Arthroplasty in Elderly People Aged 75 Years or Older

Abstract

Purpose

The study investigated the benefits of the direct anterior approach (DAA) compared to the posterolateral approach (PLA) in patients over 75 years of age.

Materials and Methods

This study included 144 patients who underwent total hip arthroplasty (THA) from December 2012 to November 2021. Group A had 93 patients with a mean age of 80.8±5.0 years, who underwent DAA. Group B had 51 patients with a mean age of 79.7±4.6 years, who underwent PLA. Clinical outcomes included operative time, time to ambulation, walking ability, and complications.

Results

There were no demographic differences between the groups. The mean age was 80.9±5.0 years in Group A and 80.5±4.8 years in Group B. Mean operative time was 94.2±7.2 minutes in Group A and 91.2±8.8 minutes in Group B (P=0.02). Early ambulation within 3 days postoperatively was seen in 72 patients (77.4%) in Group A and 31 patients (60.8%) in Group B (P=0.03). No significant change was seen in modified Koval Index in Group A (4.35 to 4.06, P=0.51), while Group B showed a significant decrease (4.47 to 3.88, P=0.04). The postoperative modified Koval index negatively correlated with time to ambulation (P=–0.17, P=0.04). Dislocation occurred in 3 patients (3.2%) in Group A and 7 patients (13.7%) in Group B (P=0.02). No differences were found in medical complications or mortality.

Conclusion

THA via DAA may provide earlier functional recovery than PLA, with comparable safety in patients over 75 years of age.

INTRODUCTION

Due to extended life expectancy, fractures of the peripheral hip joint structure due to osteoporosis and degenerative arthritis are increasing in the elderly. In particular, proximal femur fractures occurring in the elderly are becoming a social problem due to the economic burden of serious health care issues. Most hip fractures and arthritis in the elderly require surgical treatment and, currently, hip arthroplasty is performed with high frequency^1-3). Therefore, there is a need to conduct a comparative study on the degree of complications and recovery from total hip arthroplasty (THA) surgical approaches in elderly patients.

Pneumonia occurs more frequently in elderly patients requiring THA⁴⁾. In particular, aspiration pneumonia is one of the uncommon but fatal complications in elderly patients after hip arthroplasty. Aspiration pneumonia is known to be particularly related to the supine position. The elevated position (upright position and/or with head up) is known to reduce the likelihood of aspiration as compared to the supine position^5,6). It is known that postoperative delirium in the elderly with hip fractures can affect the occurrence pneumonia and aspiration pneumonia⁷⁾. Venous thromboembolism (VTE) following THA is a common complication. Pulmonary thromboembolism (PTE) has been reported in approximately 1% of patients and 0.1%-0.3% died from a fatal PTE⁸⁾. There are reports that anti-coagulants can affect the VTE and PTE occurrence in hip fracture patients, and that the occurrence in high-risk groups can be reduced with appropriate use of prophylactic measures⁹⁾.

In elderly patients who require hip arthroplasty, it is necessary to minimize muscle damage during the operation to reduce pain. It is thought that if early elevated position and early ambulation are achieved by minimizing soft tissue damage during the hip arthroplasty, medical complications such as pneumonia can be prevented. In particular, it is thought that better clinical results can be obtained in elderly patients with a high risk of postoperative complications^10,11). There are various surgical approaches for THA, and each approach with its own advantages and disadvantages. In the posterolateral approach (PLA), a wide field of view is possibly accessible at the acetabular and femoral regions. However, as a result of soft tissue weakening from damage to the gluteus and hip external rotator muscles, there is also a postoperative risk of dislocation. Compared to PLA, direct anterior approach (DAA) is a dependable intermuscular approach that can minimize muscle damage, lower the risk of dislocation, and hasten recovery time, which results in the advantage of early ambulation¹²⁾. Early ambulation is a technique of postoperative care that involves a patient getting out of bed and walking as soon as possible after surgery. This helps reduce the risk of complications such as pneumonia, PTE and VTE¹³⁾. Of course, a higher number of periprosthetic fractures have been reported as compared to other approaches, requiring caution in older patients with osteoporosis. Also, a relatively extended surgical time and large amount of bleeding during the learning curve remain a problem to be overcome^14-22).

Based on the existing age criteria of 65 years or older, there is a limitation to accurately reflecting surgical impact in elderly patients due to extended life expectancy. Consequently, this study compared two groups of elderly patients aged 75 years or older according to surgical approach and investigated the limitations and benefits of these THA surgical approaches. First, compared to the PLA group that implemented, the DAA group had earlier weight-bearing by minimizing muscle damage. Second, the DAA group freely assumed the elevated position due to the relatively low concern regarding postoperative posterior dislocation, which resulted in less frequent complications such as pneumonia, PTE, and VTE due to hospital bed stabilization. Third, in the DAA group, the postoperative mortality rate was high due to extended surgical time and increased amount of bleeding.

MATERIALS AND METHODS

This study was approved by the Kwangju Christian Hospital Institutional Review Board (IRB) (approval No. KCHIRB-M-2024-024). The written informed consent was waived by the IRB due to the retrospective nature of the study.

This is a retrospective study of 176 total patients aged 75 or older who underwent THA in a single institution under a single surgeon between December 2012 and November 2021. Of the 176 patients who had never undergone hip arthroplasty before, 144 patients were completed follow-up assessments for 2 years after the operation. These patients were classified into Group A (DAA with 93 people) or Group B (PLA with 51 people) after undergoing either the direct anterior and posterolateral THA approach. With the exception of 14 patients who were lost during the follow-up period, nine patients died, and nine patients underwent lower extremity operations affecting the biomechanics of the hip joint. There was no demographic difference between the two groups. Of a total of 144 patients, femoral neck fractures constituted 119 patients, while 14 patients were due to avascular necrosis of the femoral head and 11 patients resulted from degenerative arthritis. In Group A, each patient was laid down on a specialized operating table (Hana Orthopedic Surgery Table^®) in a supine position. The modified Gibson’s PLA was utilized for Group B after placing them in a lateral decubitus position. Cementless prostheses (Corentec^®, Zimmer^®) were used for both groups. The femoral head size used during surgery was measured, and the position of the cup was measured to see if it was within the normal range through anteversion and inclination.

For femoral head size, 36 mm was generally used for stability. For patients with small femoral head sizes, 32 mm was used. The drainage tube (Hemovac^®) was removed two days after surgery in all patients. There were no differences in rehabilitation and pain control between the two groups. In all patients, postoperative rehabilitation was performed as soon as possible, with only bedside standing and walker-assisted ambulation used for weight-bearing. During bed rest, it was recommended for Group B to use the beach chair position in which hip flexion did not occur more than 90°. For bedrest in Group A, the raised position was freely utilized immediately after surgery. Pain medications, including NSAIDs (nonsteroidal anti-inflammatory drugs), acetaminophen, and tramadol, were used to control pain appropriately considering the patient’s underlying disease and condition without difference between the two groups. Outpatient follow-up was performed at 1-month intervals up to three months after surgery to investigate radiography and clinical results, and progress was confirmed every six months after one-year postoperation. The mean follow-up period was 35.2 months, ranging from 24 to 57 months. The mean time period from diagnosis to surgery was measured. As a preoperative evaluation criterion among the two groups, preoperative walking status was identified using the modified Koval index through patient and guardian interviews²¹⁾. As evaluation criteria during the operation, time span, blood loss, and the periprosthetic fracture incidence were checked. Total blood loss was calculated as the sum of the amount generated during the operation and the amount measured from the postoperative drainage tube (Hemovac^®). Clinical and radiological evaluations were performed postoperatively. Clinical assessments included the percentage of weight-bearing achieved within 3 days after surgery, the number of days until the initial weight-bearing after surgery, the mean length of hospital stay, and postoperative walking status.

Considering the advanced age of the patients, postoperative walking status was based on the highest score among the patients’ condition using the modified Koval index until two years after surgery. Medical complications, such as aspiration pneumonia, VTE, and PTE, were diagnosed by chest computed tomography and angiographic computed tomography if abnormalities, such as decreased saturation or elevated D-dimer, were detected during the hospital stay. Patients with symptoms of postoperative delirium were identified. For patients on an anticoagulant regimen, the anticoagulant was discontinued for a set number of days before surgery in consult with the cardiovascular and anesthesiology department. The postoperative bleeding condition was then assessed, and the drainage tube (Hemovac^®) was removed approximately two days after surgery. Subsequently, the anticoagulant was readministered. Mortality rates were followed within two years of surgery. Short-term mortality was categorized as occurring within six months of surgery Long-term mortality was categorized as occurring after more than six months after surgery. All causes of mortality were identified. Simple radiography evaluation measured acetabular inclination and anteversion after surgery. Complications, such as periprosthetic fracture, dislocation, loosening, subsidence, and infection, were investigated by simple radiography during follow-up. For statistical analysis, the independent t-test, Pearson’s linear correlation coefficient, and chi-square test were employed using PASW Statistics 18.0.0 for Windows software (IBM Corp.), and were considered statistically significant when the P-value was <0.05.

RESULTS

The mean age was 80.9±5.0 years in Group A and 80.5±4.8 years in Group B. Group A was sex-categorized as 15 males and 78 females, while Group B had 11 males and 40 females. The mean period from diagnosis to surgery was 4.22±2.57 days in Group A and 5.11±3.11 days in Group B, with no significant difference between the two groups (P=0.65). In Group A, the diameter of the femoral head was 36 mm in 80 patients (86.0%) and 32 mm in 13 patients (14.0%), while in Group B, 36 mm in 42 patients (82.4%) and 32 mm in 9 patients (17.6%) were used. Cup anteversion, averaging 19.62°±7.01° in Group A and 19.91°±1.63° in Group B, was within the normal range for both groups. Inclination, 42.51°±6.32° in Group A and 44.90°±2.73° in Group B, was within the normal range for both groups. The mean amount of blood loss during operation was 429.2±88.2 mL in Group A and 424.0±113.1 mL in Group B. Although there was relatively higher amounts of blood loss in Group A, there was no significant difference (P=0.77) between groups. Mean operational time was longer in Group A (94.2±7.2 minutes) (P=0.02) than in Group B (91.2±8.8 minutes). Periprosthetic fracture did not occur in either group during surgery. Acetabular inclination and anteversion were compared through simple radiography immediately after surgery, and no significant difference between the two groups was found (Table 1). In Group A, 72 patients (77.4%) were able to walk with assisted walking weight-bearing within three days of surgery, and 21 patients (22.6%) achieved assisted weight-bearing after three days. In Group B, 31 patients (60.8%) were able to walk within three days and 20 patients (39.2%) after three days. The proportion of patients who could walk with assisted walking weight-bearing within three days was smaller than in Group A, and there was a statistically significant difference (P=0.03). The mean number of days for patients’ postoperative assisted walking ambulation in Group A was 4.8±6.5 days and 7.8±9.2 days in Group B. There was also a statistically significant difference in the number of days to initiate walking in Group A (P=0.03) (Table 1). When measured for both groups, a modified Koval index showed a decrease from a mean of 4.40±1.21 before surgery to 4.00±1.33 after, but it was not significant (P=0.28). When measuring Group A, it decreased from a mean of 4.35±0.98 to a mean of 4.06±0.87, but this was not significant (P=0.51). When measuring Group B, the mean decrease was more than that of Group A, from 4.47±1.12 to 3.88±1.57, and showed a significant decrease (P=0.04) (Fig. 1). During follow-up, the correlation of the modified Koval index and the timing of walking assistance after surgery was observed, and the numerical value of the modified Koval index decreased as the walking assistance timing was delayed, and it was significant (R=–0.17, P=0.04) (Fig. 2). During the postoperative hospital stay, the incidence of aspiration pneumonia according to surgical approach occurred in 3.2% of patients (3/93) in Group A and 7.8% of patients (4/51) in Group B. No significant difference was found. In Group A, postoperative delirium occurred in 8.6% of patients (8/93). Postoperative delirium occurred in 9.8% of patients (5/51) in Group B. Although the incidence was higher in Group B, there was no significant difference. While PTE occurred in 2.2% of patients (2/93) in Group A and in 7.8% of patients (4/51) in Group B, no significant difference was found between groups. VTE occurred in 2.2% of patients (2/93) in Group A and 3.9% of patients (2/51) in Group B, and no significant difference found between groups. Of the patients on anticoagulant therapy, 51.6% (48/93) were in Group A and 49.0% (25/51) were in Group B. In Group A, complications in the anticoagulant therapy subset of patients presented as PTE in 2.1% of patients (1/48) and as VTE in 0.0% of patients (0/48). The complications of patients on anticoagulant therapy in Group B presented as PTE in 4.0% of patients (1/25) and VTE in 4.0% of patients (1/25). No significant difference was found between groups. The short-term mortality rate within six months of Group A was 2.2% (2/93), the long-term mortality rate (beyond six months) was 3.2% (3/93). Short-term mortality in Group B was 2.0% of patients (1/51) and long-term mortality was 5.9% of patients (3/51). All-cause mortality occurred in 5.4% of patients (5/93) in Group A and 7.8% of patients (4/51) in Group B. There was no significant difference in mortality within two years according to the surgical approach. Three dislocations (3.2%) occurred in Group A, and seven dislocations (13.7%) occurred in Group B. As such, there was a statistically significant difference (P=0.02) (Table 1). During follow-up, no periprosthetic fractures, loosening, subsidence, or infection were observed in either group.

Table 1.

Comparison of Perioperative Results, Rehabilitation, and Complications between DAA and PLA

	Group A	Group B	χ² value	P-value
Perioperative results
Bleeding (mL)	429.2±88.2	424.0±113.1	-	0.77
Operative time (min)	94.2±7.2	91.2±8.8	-	0.02*
Periprosthetic fracture	0/98 (0.0)	0/55 (0.0)	-	-
Acetabular inclination (°)	42.51±6.32	44.90±2.73	-	0.34
Acetabular anteversion (°)	19.62±7.01	19.91±1.63	-	0.83
Rehabilitation
In 3 days	72/93 (77.4)	31/51 (60.8)	4.5	0.03*
After 3 days	21/93 (22.6)	20/51 (39.2)
Days for first weight-bearing	4.8±6.5	7.8±9.2	-	0.03*
Complications
Aspiration pneumonia	3/93 (3.2)	4/51 (7.8)	1.5	0.22
Delirium	8/93 (8.6)	5/51 (9.8)	2.1	0.25
VTE	2/93 (2.2)	2/51 (3.9)	0.4	0.54
PTE	2/93 (2.2)	4/51 (7.8)	2.7	0.10
All-cause mortality	5/93 (5.4)	4/51 (7.8)	0.3	0.72
Dislocation	3/93 (3.2)	7/51 (13.7)	-	0.02*

Values are presented as mean±standard deviation or number (%).

Group A: DAA, Group B: PLA.

DAA: direct anterior approach, PLA: posterolateral approach, VTE: venous thromboembolism, PTE: pulmonary thromboembolism.

Statistics by independent t-test or chi-square test; *P<0.05.

Fig. 1.

Comparative graph of changes in modified Koval index in Group A (DAA) and Group B (PLA). DAA: direct anterior approach, PLA: posterolateral approach, Fx.: fracture. Statistics by independent t-test; *P<0.05.

Fig. 2.

Spotted graph of correlation analysis between number of days for first weight-bearing and modified Koval index after total hip arthroplasty (THA) in all patient groups (R=–0.17, P=0.04). Statistics by Pearson’s linear correlation coefficient and chi-square test.

DISCUSSION

Most existing THA studies in the elderly have compared PLA and DAA by applying chronological age criteria of 65 years or older^21,22). Considering complications that may occur after THA in an aging society, it is important to evaluate how early postoperative ambulation affects complication occurrence and walking ability in late elderly people older than 65 years of age. Accordingly, this study was unprecedented as it was conducted on the elderly older than 65 years of age. Empirically, it statistically confirmed the importance of early weight-bearing. In this study, there were a significant number of patients weight-bearing within three days in Group A, and the first walking-assisted ambulation occurred significantly sooner in Group A. In addition, within two years, modified Koval index after THA was high in Group A with no significant decrease before or after surgery. In the case of DAA, it was thought that it was possible to reduce postoperative pain due to less muscle damage. Additionally, because there is less fear of posterior dislocation, the patient can posture more freely in bed after surgery. For this reason, it is thought that it can give patients a sense of stability, make it easier to get out of bed, and allow them to start walking more quickly. This is corresponded with the Ang et al.²¹⁾ study which reported that there was further benefit to functional recovery in the early postoperative period and that DAA patients had a shorter hospital stay than PLA patients. Our study statistically analyzed the correlation between walker assisted ambulation and the modified Koval index after surgery. When initial weight-bearing time was delayed, the modified Koval index showed a significant negative correlation. This indicated that the amount of walking after surgery might be greater in Group A, which had earlier weight-bearing and earlier walking time, than in Group B. In particular, the frequency of weight-bearing within three days after surgery was significantly high in Group A, which reaffirmed the usefulness of DAA in terms of functionality.

Although DAA has multiple postoperative advantages in elderly patients, concerns have been reported about prolonged surgical time and complications²²⁾. In this study, there was no significant difference in age and sex distribution. Although mean surgical time was longer in DAA, there was not much difference in mean time between the THA approaches studied. Femoral head size and cup position were used as important parameters for determining dislocation. The majority of cases in the two groups used a femoral head of 36 mm, and the cup position was within the normal range in both groups, with no significant difference. The amount of blood loss was higher in Group A than in Group B. However, the difference was not significant. In most other perioperative complications, there was no significant difference between PLA and DAA in patients over 75 years of age. In the elderly over 75 years of age, medical complications after hip surgery are affected by various factors. Therefore, it is limited to confirm that only elevated bed position and early weight-bearing are correlated between the prevalence and mortality of the above diseases. According to Kim et al.²⁰⁾, the incidence of VTE and PTE is high within 24 hours after hip fracture surgery, and the effect of appropriate anticoagulant therapy during this period is positive⁹⁾. However, patients in this study sufficiently discontinued anti-coagulant therapy and then subsequently resumed it two days after surgery on mean; it is possible that it did not significantly affect the incidence of VTE and PTE.

In follow-up radiographic results, dislocation was significantly more common in Group B than in Group A. These results implied that there was a small amount of soft tissue damage and less occurrence of dislocations in the DAA cases due to the absence of the inevitable gluteal and short external rotator muscle injury that occurred in PLA cases. Periprosthetic fractures, loosening, subsidence, and infection were not observed in either group. Consequently, it was found that there were no significant differences between the two groups in the occurrence of medical complications, bleeding volume, mortality, periprosthetic fractures, loosening, subsidence, and infection. The extent of walking ability and hip dislocation after surgery were significantly better off in Group A. This result indicates that DAA can be a safe equivalent to PLA in older patients.

Limitations of this study include: First, a small number of samples of DAA and PLA were used for this study. Second, this was a retrospective study utilizing date from a single surgeon in a single institution. Third, this was a limited evaluation of surgical approaches and early postoperative walking without considering various risk factors associated with aspiration pneumonia, VTE, and PTE. Understanding of those risk factors should be considered through future research studies.

CONCLUSION

In elderly patients over the age of 75 years, THA with DAA was determined to be as safe as THA with PLA. Specifically, the results of our study revealed that DAA is one of the most useful surgical approaches in the elderly because early weight-bearing can be achieved by minimizing soft tissue damage, and DAA patients were able to freely assume an elevated position compared to others who underwent PLA.

Funding Statement

Funding No funding to declare.