The potential benefits of minimally invasive total hip arthroplasty
continue to be heavily debated by the orthopaedic community. The definition of
minimally invasive surgery is somewhat ambiguous at present, but, in general,
it applies to the performance of the procedure through a smaller incision or
an alternative surgical approach with less tissue
dissection1-5.
Studies have described a better outcome after total hip arthroplasty
performed both through a smaller
incision6-9
and through an alternative approach such as the two-incision
technique1,8
or an anterior approach
5,10,11.
The advocates of minimally invasive total hip arthroplasty have cited less
pain, quicker recovery, better patient satisfaction, and less blood loss as
some of the advantages of this
procedure1,6,7,9,12,
whereas other surgeons have refuted its
superiority13-19.
Furthermore, studies have shown a worse cosmetic appearance of the
wound20 or an
increased degree of abductor muscle damage with minimally invasive total hip
arthroplasty21.
It is uncertain whether a minimally invasive approach confers any benefits
to patients and, if it does, what factors are indeed responsible for the
improved outcome that has been reported by some
investigators6-9,12,22.
It is possible that confounding factors, such as patient selection, patient
and family education, accelerated rehabilitation, and better pain control, may
play an important role in influencing the outcome of minimally invasive total
hip
arthroplasty23.
This study was designed to investigate the role of these factors in general
and accelerated rehabilitation in particular on the outcome of total hip
arthroplasty.
Demographic Data
Following institutional review board approval of the study, consecutive
patients undergoing unilateral total hip arthroplasty performed by two
surgeons (R.H.R. and W.J.H.) at our institution during an eight-month period
were approached and evaluated for inclusion in this study. The inclusion
criteria were an age of between eighteen and seventy-five years, any gender or
race, an underlying diagnosis of osteoarthritis, and agreement to provide
consent to participate in the study. Patients with a body mass index of >30
kg/m2 or those with cognitive impairment or severe psychiatric
illness that would preclude participation in the protocol-mandated procedures
were excluded. The purpose of the study and the method, including the
randomization, were discussed in detail with the patient before consent was
obtained. One hundred patients were enrolled. During the period of the study,
fifty-seven other patients were also screened for inclusion. Fifteen patients
refused to participate, eighteen had a body mass index of >30
kg/m2, twelve were more than seventy-five years of age, and twelve
had a diagnosis other than osteoarthritis. The enrolled patients were then
randomized, with use of computer-generated cards, into one of four groups.
Group A consisted of patients who underwent total hip arthroplasty through a
standard incision (a length of >10 cm) and received standard preoperative
and postoperative care. Group-B patients underwent total hip arthroplasty with
use of a small incision (=10 cm) and were subjected to standard
preoperative and postoperative protocols. Group-C patients underwent a total
hip arthroplasty with a standard incision but received preoperative
counseling, accelerated rehabilitation, and an altered pain control regimen.
Group-D patients underwent total hip arthroplasty through a small incision and
received preoperative counseling, accelerated rehabilitation, and an altered
pain control regimen. In six patients in the small-incision groups (two in
group B and four in group D), extension of the incision was required at the
time of surgery because of difficulty with visualization. These six patients
were excluded from the final analyses. The demographic distribution of the
ninety-four patients who constituted the study group is shown in
Table I.
Surgical Data
All patients received spinal anesthesia. Approximately 0.2 mg of
DepoMorphine (morphine) was administered intrathecally at the time of spinal
anesthesia for all patients. Intravenous propofol, midazolam, and narcotic
analgesics such as morphine, fentanyl, or meperidine were administered at the
discretion of the anesthesiologist. A detailed record of all intraoperative
analgesics was obtained.
Patients were positioned supine with a flat bump under the back. The skin
incision was based on the greater trochanter. The length of the incision
averaged 9.6 cm (range, 7 to 10 cm) for the small-incision group and 13.8 cm
(range, 11 to 19 cm) for the standard-incision group (p < 0.0001). The
fascia lata was divided in line with the incision, and the anterior third of
the gluteus medius and gluteus minimus muscles were reflected off the greater
trochanter. The dissection was carried distally to include 2 to 3 cm of the
proximal portion of the vastus lateralis. An anterosuperior capsulotomy was
performed to expose the hip joint.
An uncemented total hip arthroplasty with use of a proximally coated
tapered stem (Accolade; Stryker Orthopaedics, Mahwah, New Jersey) and a
plasma-sprayed acetabular component (Trident; Stryker) was performed in all
patients. The type of bearing surface utilized was at the discretion of the
surgeons and depended on the age of the patient. The preference of the
surgeons was to use a ceramic-on-ceramic bearing for patients younger than
sixty years. A ceramic femoral head articulating with a polyethylene
acetabular liner was typically used for patients between the ages of sixty and
seventy years. Patients older than seventy years received a
metal-on-polyethylene bearing surface.
Closure of the deeper tissue was carried out with the use of a braided
absorbable suture (number-1 Vicryl [polyglactin]; Ethicon, Somerville, New
Jersey). The fascia lata and the subcutaneous tissue were also closed with
absorbable sutures in an interrupted manner, and the skin was closed with
staples. No surgical drains were used.
The only difference between the small-incision and standard-incision groups
regarding the operative technique was the size of the skin and subcutaneous
tissue incision (=10 cm compared with >10 cm). The deep dissection did
not vary.
The details of all surgical parameters, including estimated blood loss,
operative time, intraoperative complications, and any other pertinent
information, were recorded.
Preoperative Protocols
Standard Protocol
Patients were seen in the outpatient office alone or together with a family
member. They were provided information on total hip arthroplasty. They were
informed that they would be placed on patient-controlled analgesia
postoperatively for two to three days. The postoperative rehabilitation was
discussed, and patients were told that walking with assistance on a daily
basis would commence the day after surgery. They were also told that they
would either be discharged home or to a skilled rehabilitation facility three
or four days after surgery. Patients received no preemptive analgesia. They
were told to discontinue the use of nonsteroidal anti-inflammatory
medications, if any, two weeks prior to the surgery. Incision size was not
discussed.
Accelerated Protocol
The patients were evaluated in the outpatient office together with a family
member. The patients were told that they would be subjected to accelerated or
"fast-track" surgery. The postoperative analgesia was described as
consisting of oral analgesics postoperatively supplemented by intravenous
medications, if needed. The patients were reassured that their pain would be
controlled well. The patients were told about the benefits of walking early
and were encouraged to comply with the rehabilitation protocol that would
involve assisted walking on the day of surgery and twice daily thereafter. The
patient and family member were told that home discharge would be based on the
ability of the patient to walk independently and climb stairs, which would
most likely occur on the second postoperative day. Patients were told that
discharge to home would be preferred, and family members were encouraged to
care for the patient at home. Patients were seen by a physical therapist for
two or three sessions preoperatively for gait training and familiarization
with postoperative exercises. Patients were placed on a selective
cyclooxygenase-2 inhibitor medication (200 mg of celecoxib once daily)
starting at least seven days prior to surgery. Those randomized to the
small-incision group were also told that they would undergo surgery with use
of a "mini-incision" technique.
Postoperative Management
Appropriate prophylaxis against infection (Ancef [cefazolin] preoperatively
and for twenty-four hours after the surgery) and thromboembolism (Coumadin
[warfarin] for six weeks) was administered to all of the patients according to
our standard protocol.
Blood management included readministration of one unit of predonated blood
to all patients who had donated pre-operatively. Allogenic transfusion was
carried out for patients with a hemoglobin level of <8 g/dL or for
symptomatic patients with a hemoglobin level of between 8 and 10 g/dL.
Symptoms triggering transfusion included persistent tachycardia (a heart rate
of >100 for at least four hours), lassitude (an inability to comply with
physical therapy exercises), chest pain, dyspnea, and hypotension (a decrease
in blood pressure of >20 mm/Hg). The internal medicine specialists, unaware
of the status of the patients in the study, were responsible for the
management of blood replacement. The details of blood transfusion were
recorded.
All patients received a social service consultation to discuss social
circumstances, and the need for disposition to a skilled rehabilitation
facility was determined on the basis of the degree of home support, the layout
of the home, and the physical ability of the patient. On occasion, patients
who were initially assumed to be candidates for home discharge were sent to a
skilled rehabilitation facility at the recommendation of the physical therapy
team.
The cumulative dose and the type of analgesic medications administered
during the hospital stay and for up to six weeks postoperatively were recorded
in detail for each patient. The equianalgesic dose for all drugs was
calculated. The standard equianalgesic conversion based on pain-relieving
efficacy equivalent to 10 mg of morphine was
used24-26.
There were some inherent differences in rehabilitation and pain control
between the standard and accelerated groups:
Standard Group
Patients were placed on intravenous patient-controlled analgesia usually
containing morphine, which was initiated postoperatively in the recovery area,
and they continued to use it for a minimum of two days. The standard dose (1
mg of morphine with a six-minute lock-out and a maximum of 10 mg/hr) was
administered at first. Depending on the degree of pain and patient
satisfaction with pain control, other analgesics including fentanyl were also
administered ad libitum or the setting for the patient-controlled analgesia
and/or the type of narcotic used was altered. Supplemental analgesia was
provided if necessary. Transition from patient-controlled analgesia to oral
medications (OxyContin [oxycodone], Lortab [hydrocodone], or Percocet
[oxycodone]) occurred on the third postoperative day.
Patients were seen by a physical therapist on the first postoperative day
and were assisted out of bed and into a chair. They were also encouraged to
walk, if they were able. Sitting in a chair at least three times a day and
assisted walking once a day were carried out thereafter. Patients were
evaluated by the physical therapist to determine their potential for
stair-climbing on the third postoperative day. Patients were sent home or to a
skilled rehabilitation facility depending on their walking status and the
circumstances at home.
Accelerated Group
Patients in this group were placed on an oral narcotic medication (5 mg of
OxyContin [oxycodone] every four to six hours as needed) and, if they were
able to tolerate it, a selective cyclooxygenase-2 inhibitor medication (200 mg
of celecoxib once daily) on the day of surgery. Supplemental intravenous
medication was also provided ad libitum. Patients had also received preemptive
analgesia for at least seven days prior to surgery.
The patients were seen by a physical therapist a few hours after arrival on
the ward and were helped to sit in a chair or to walk with assistance if
possible. Physical therapy occurred at least twice daily thereafter.
Functional Outcome
Preoperative and early postoperative functional outcome was measured with
use of linear analog scale
assessment27, the
Western Ontario and McMaster Universities Osteoarthritis Index
(WOMAC)28, the
Short Form-36
(SF-36)29, the
Harris hip score30,
and the lower extremity functional
score31. Functional
assessment on a daily basis in the hospital was also carried out with use of
the linear analog scale assessment. Data on the patients with regard to
walking status (able to walk independently or walk with supervision, minimal
assistance, or contact guard), walking distance, and type of walking aid used
(single-point cane, axillary crutches, rolling walker, or standard walker)
were recorded and compared among the groups.
Follow-up
At our institution, the total joint registry collects clinical and
radiographic data on all arthroplasty patients prospectively. A strict
follow-up protocol was in place to closely monitor the patients recruited into
this study. All patients were seen in the office and were examined by the
senior surgeons (R.H.R., W.J.H. and P.F.S.) at or around six weeks
postoperatively. Clinical examination and radiographic evaluation were
performed during the follow-up visit.
Statistical Analysis
Prior to the initiation of the study, a power analysis was performed with
the assistance of a statistician to ensure that adequate sample size was
selected. For an effect size of 20% in early functional outcome, as measured
with a validated instrument such as the linear analog scale assessment, with
a = 0.05 and ß = 0.80, calculations revealed that forty patients in
the mini-incision group and forty patients in the standard-incision group
would be needed. In addition to the required number of subjects, twenty more
patients were recruited to allow for possible attrition. The unpaired
independent t test was used to compare the mean with variances among all study
outcome parameters. The chi-square test and the Fisher exact test were used to
compare the nonparametric outcomes. The level of significance was set at p
< 0.05.
Functional Outcome
Following total hip arthroplasty, there was a significant improvement in
function as measured by the Harris hip score, the linear analog scale
assessment, the SF-36, the WOMAC, and the lower extremity function test in all
four groups. Patients who were subjected to the accelerated rehabilitation
protocol regardless of the size of the incision had a significantly better
outcome with respect to the postoperative mental health element (p = 0.03) of
the SF-36 and selected components of the linear analog scale assessment
(energy level, p = 0.01; daily activity, p = 0.007; and quality of life, p =
0.02) (Table II).
Patients in the accelerated rehabilitation groups could walk significantly
longer distances at the time of discharge compared with those in the standard
physical therapy groups (35 and 24 m, respectively; p = 0.008). A
significantly higher percentage of patients in the accelerated physical
therapy groups could walk independently or with minimal supervision at the
time of discharge (p = 0.009) (Table
II). The length of the incision did not seem to influence any of
these parameters (Table
III).
Analgesia Requirement
There was no difference in analgesic requirement between the small-incision
and standard-incision groups (p = 0.65). There was, however, a significant
difference between the standard and accelerated physical therapy groups with
regard to narcotic consumption (p = 0.01). Patients undergoing accelerated
rehabilitation consumed more narcotic medication because of an increase in the
use of oral narcotics such as OxyContin (oxycodone).
Blood Requirements
The patients with a small incision and those with a standard incision did
not differ with respect to the estimated blood loss (p = 0.19), decrease in
hemoglobin level (p = 0.89), and need for transfusion (p = 0.7)
(Table III).
Length of Hospital Stay
The mean length of hospital stay for the patients managed with the
accelerated rehabilitation protocol (3.5 days; range, two to five days) was
significantly shorter than that for the patients managed with the standard
protocols (4.2 days; range, three to eight days) (p = 0.001). More patients
were also discharged home, rather than to a skilled rehabilitation facility,
in the accelerated rehabilitation group (p = 0.02)
(Table II). The length of
hospital stay for patients with a small incision and those with a standard
incision was not significantly different (p = 0.5).
While the debate regarding the potential benefits of minimally invasive
total hip arthroplasty continues, the influence of a variety of factors on the
outcome of total hip arthroplasty is not completely known. In a recent survey
of members of the Hip Society in 2004, 77% of the forty-five surgeons
confirmed that they had implemented changes in their routine practice when
performing minimally invasive total hip
arthroplasty32.
These included the use of preemptive analgesia, preoperative patient
education, changes in anesthesia, the use of different surgical instruments,
and the use of an aggressive rehabilitation
protocol33.
The present study was conceived to evaluate the influence of family
education, patient conditioning, preemptive analgesia, improved postoperative
pain control, and accelerated rehabilitation on the outcome of total hip
arthroplasty. We found that patients receiving accelerated physical therapy
and improved pain control demonstrated better improvement in function, had
better satisfaction, and were discharged earlier than those who received
standard rehabilitation and postoperative analgesia. The size of the incision
did not seem to affect these measured outcomes. The study highlights the fact
that implementation of changes in surgical and anesthesia protocols, and not
the surgical technique per se, may play a major role in creating the better
outcome with a minimally invasive total hip arthroplasty that has been
reported by various
investigators1,2,6,8,12,22.
Hence, the introduction of the minimally invasive total hip arthroplasty
concept should be applauded as it has brought positive changes in
rehabilitation and pain control to the conventional practice of total hip
replacement.
Some caveats need to be borne in mind when interpreting the findings of
this study. First, the patients undergoing total hip arthroplasty in this
study all had the procedure through exactly the same surgical approach and
only the size of the incision differed. Hence, by strict definition, this was
a comparison between a small incision and a standard incision and was not
between minimally invasive total hip arthroplasty and conventional total hip
arthroplasty. Nevertheless, total hip arthroplasty performed through a small
incision has been labeled as minimally invasive total hip arthroplasty by some
investigators who report advantageous outcomes of this
approach1,6,7,9,17.
In addition, in a recent survey, two-thirds of the nineteen surgeons who said
they performed minimally invasive total hip arthroplasty defined it as
small-incision surgery, while the other third performed it through an
alternative tissue plane (with two incisions or using an anterior
approach)33.
Second, another important point is that this study was conceived to evaluate
the role of multiple factors on the outcome of total hip arthroplasty at the
same time, and we did not intend to separate these confounding factors. In
other words, the influence of preoperative physical therapy, preemptive
analgesia, family and patient education, accelerated postoperative
rehabilitation, and the use of oral as opposed to intravenous analgesia
postoperatively were all being tested at once and in combination. Hence, it is
difficult to distinguish which one of these factors may have played a more
critical role in affecting the measured parameters. The design of the study in
not separating these factors was intentional. The study achieved its intended
objective by demonstrating that factors other than the length of the incision
can affect the early outcome of total hip arthroplasty.
It is not difficult to comprehend why patient education and preconditioning
could positively impact the outcome of a surgical procedure. In a recent
study, patients who received cognitive and behavioral training before an
endoscopy procedure experienced substantially fewer signs of behavioral
distress during the procedure than those who did not receive such training
before the
procedure34. The
beneficial effects of educational preparation on the outcome of total knee and
hip replacement has also been
demonstrated35,36.
The beneficial effect of preemptive administration of analgesia on
orthopaedic patients has also been demonstrated in several
studies37-42.
In addition, the efficacy of multimodal pain control protocols, consisting of
regularly administered oral analgesics with intravenous medications for
breakthrough pain, has been observed by previous
investigators37-40,42,43.
In particular, perioperative use of a cyclooxygenase-2 inhibitor has been
shown to reduce opioid consumption, pain, vomiting, and sleep disturbance,
with improved knee range of motion after total knee
arthroplasty43. On
the basis of these reports, we had elected to administer oral analgesics
supplemented by intravenous narcotic medications for breakthrough pain in
patients who were randomized to the accelerated rehabilitation group. We found
that the equianalgesic consumption of narcotic medications in that cohort
provided better patient satisfaction and improvement in function.
The findings of this study have stimulated changes in our protocols that
allow earlier walking and improved patient satisfaction without an increase in
complications. Total hip arthroplasty at our institution is now performed
through an incision that is as small as possible and yet allows safe and
secure positioning of the components without a preoccupation with its actual
size. ?