Open Journal of Orthopedics, 2011, 1, 9-13
doi:10.4236/ojo.2011.12003 Published Online December 2011 (
Copyright © 2011 SciRes. OJO
Pulsed Lavage in Cementless Total Hip
Arthroplasty Reduces the Incidence of
Brooker Grade 3 and 4 Heterotopic Ossifications*
Joris Mellema1, Wierd Zijlstra2, Tom van Raaij2, Basten Have2, Jos van Raay2
1Department of Surgery, Isala klinics, Zwolle, The Netherlands; 2Department of Orthopaedic Surgery, Martini Hospital, Groningen,
The Netherlands.
Received October 9th, 2011; revised November 16th, 2011; accepted November 24th, 2011.
Heterotopic ossification (HO) may cause pain, and can lead to loss of hip motion after total hip arthroplasty (THA).
There is evidence that pulsed lavage may lower the incidence of HO formation. We assessed the effect of pulsed lavage
on the incidence of HO in 87 male patients after THA. All patients received an uncemented THA through a poster-
olateral approach. 39 patients were treated with pulsed lavage (index group) and 48 males were treated without pulsed
lavage (historical control group, matched on aetiology, gender, surgical approach and type of prosthesis). Both groups
followed the same postoperative treatment regimen. HO severity was scored in both groups according to the Brooker
classification by three blinded orthopaedic surgeons one year postoperatively. Good inter-observer agreement (Kappa
0.7) for scoring HO was found. The incidence of HO (51%) in the index group did not differ significantly (p = 0.53)
from the control group (58%). However, the incidence of clinically relevant HO (Brooker grades 3 and 4) was signifi-
cantly lower (p = 0.04) in the index group (3%) as compared to the control group (17%). These results suggest a bene-
ficial effect of pulsed lavage on the incidence of severe heterotopic ossification after cementless THA in male patients.
Keywords: Heterotopic Ossification, Pulsed Lavage, Total Hip Arthroplasty, Uncemented
1. Introduction
Heterotopic ossification (HO) is a well-known co mplica-
tion after total hip arthroplasty (THA). The incidence of
HO varies from 8% to 90% depending on risk factors and
the criteria used [1,2]. A large meta-analysis of 13 ran-
domized trials, involving 4129 individuals treated with
Non Steroid Anti Inflammatory Drugs (NSA IDs) after
major hip surgery, reported an overall HO incidence of
37% [3]. Especially advanced stages of HO, Brooker
grades 3 and 4, are clinically relevant because of pain
and hip function impairment [4-6].
The exact aetiology of HO remains unclear. Some
have suggested that surgical trauma may stimulate pri-
mitive mesenchymal cells to differentiate into osteoblastic
cells, which form bone [7,8]. Earlier, Lo and Healy [9]
found that non-circulating connective tissue cells with
fibroblastic features may cause HO. It seems likely that
HO is the result of inappropriate differentiation of pluri-
potential mesenchymal cells into osteoblastic stem cells.
Some risk factors for HO around the hip have been
identified; male gender, a lateral surgical approach, un-
cemented arthroplasty, HO after contralateral hip arthro-
plasty and revision surgery [10,11]. Several studies re-
ported beneficial effects of post-operative NSA IDs us-
age [12] and local radiation therapy [13] in preventing
HO formation. However, NSAIDs often cause gastroin-
testinal side effects, and local radiation may generate
higher costs, logistic problems, wound healing disorders
and introduces the risk of secondary malignnancies.
Pulsed lavage around the hip jo int and gluteal muscles
may prevent HO formation by washing out the primitive
mesenchymal cells. It is inexpensive and has no known
side effects. Only one study on the effect of pulsed lav-
age has been published and suggested no protective ef-
fect on HO formation [1]. In this study, however, all pa-
tients received lavage; pulsed or manual with a syringe.
Aim of the current study was to determine whether pul sed
lavage compared to no lavage resulted in less clinically
relevant HO formation after cementless THA in male
*Conflict of Interests: None declared
Pulsed Lavage in Cementless Total Hip Arthroplasty Reduces the Incidence of Brooker
10 Grade 3 and 4 Heterotopic Ossifications
2. Materials and Methods
Between May 2003 and February 2006, 41 consecutive
male patients scheduled for cementless THA were in-
cluded prospectively. All patients underwent cementless
THA through a posterolateral approach because of pri-
mary hypertrophic coxarthrosis, and received pulsed
lavage during the procedure. Two patients were lost to
follow-up. The index group of 39 patients was matched
(on aetiology, gender, surgical approach, type of prosthe-
sis and time of follow-up) to a historical control group of
48 patients who received a cementless THA without any
form of lavage between March 1993 and December 2001.
The local medical ethical committee approved the study
design. Informed consent was obtained in all patients.
The trial was performed in compliance with the Helsinki
Declaration. There were no competing interests.
2.1. Surgical Procedure
All patients received a unilateral plasma sprayed porous
coated titanium alloy ce mentless total hip pros thesis with
a polyethylene liner (Mallory Head®, Biomet, Warsaw,
Indiana, USA) through a standard posterolateral ap-
proach. No lavage or irrigation was used in the control
group. The index group received pulsed lavage with a
total of 1 litre saline at 2 time points; 1:50 0 mL after im-
plantation of cup and stem with the hip in dislocated po-
sition and 2:500 mL after reposition. Pulsed lavage was
performed with the OptiLavage system® (Biomet, War-
saw, Indiana, USA) in and around the joint and gluteal
muscles. After surgery a suction drain was used in all
patients routinely. According to protocol no NSAIDs
were given postoperatively in both groups. The rehabili-
tation programme was similar for both groups, and con-
sisted of full weight bearing from the first day after sur-
2.2. Scoring Procedure
HO formation was scored using the Brooker classifica-
tion on standardized supine anteroposterior pelvic radio-
graphs at time of follow-up [14]. Three blinded orthopae-
dic surgeons scor ed all radiographs to assess inter-obser-
ver variability [5]. Th e mode of the 3 observers was used
as definitive Brooker score when disagreement in HO
rating oc curred.
2.3. Statistical Analysis
Baseline characteristics of both groups are displayed us-
ing descriptive statistics. Differences between groups
were tested with Student’s T-test or Mann-Whitney U-
test in case of non-normal distribution of a parametric
variable or non-p arametric var iable. In ter-observ er agree-
ment between the 3 observers in scoring of HO formation
using the Brooker classification was determined with the
kappa statistics. Numbers and percentages of the amount
of HO formation in both grou ps we re calculated. Brooker
scores were divided into 2 groups; grades 1 and 2 (clini-
cally irrelevant) and grades 3 and 4 (clinically relevant)
[4-6]. Differences between both groups in Brooker gr ades
were tested with the Fischer’s Exact test. All analyses
were performed with SPSS version 16.0 for Windows
(SPSS inc., Chicago, Illinois). A p value < 0.05 was con-
sidered statistically significant.
3. Results
Both groups were equal regarding aetiology, gender, sur-
gical approach, type of prosthesis, and time of follow-up.
The index group was significantly (p = 0.002) older with
a mean age of 62 years as compared to the control group
mean age of 55 years (Table 1).
Good inter-observer agreement (K = 0.7) for scoring
HO formation according to the Brooker classification
was found (Table 2).
Figure 1 shows the distribution of Brooker grades in
the index and control group. HO formation was found in
20 out of 39 patients in th e index group, and in 28 out of
48 patients in the control group. Total HO incidence
percentage (51% and 58%, respectively) did not differ
significantly (p = 0.53) between both groups. However,
the index group (1 of 39 patients) had significantly (p =
0.04) less Brooker grades 3 and 4 HO compared to the
control group (8 of 48 patients); 3% versus 17%, respec-
Table 1. Baseline characteristics for the index and control
Group Index(N = 39) Control (N = 48)
Follow up, months (mean, SD)12 (7) 15 (8)
Age, years (mean, SD) 62 (9)ª 55 (10)ª
ap = 0.002.
Table 2. Inter-observer variability for scoring HO forma-
tion (Brooker classification).
Observer combination ĸ coefficient
O1-O2 0.779
O1-O3 0.654
O2-O3 0.734
Average 0.722
Copyright © 2011 SciRes. OJO
Pulsed Lavage in Cementless Total Hip Arthroplasty Reduces the Incidence of Brooker 11
Grade 3 and 4 Heterotopic Ossifications
Figure 1. Distribution of the Brooker grades between the
index and control groups.
4. Discussion
Our results demonstrated significantly less severe HO
formation (Brooker grades 3 and 4) with the use of puls ed
lavage in male patients treated for symptomatic primary
hypertrophic coxarthrosis with uncemented THA. Given
the large variation in HO incidence from 8% - 90%
across different trials [1,2], and the use of a historical
control group, rather than a randomised study, we con-
sider that the low numbers compared at Brooker grade 3
& 4 could lead to misleading conclusions. The present
study was limited that patients were not randomized,
which may prevent correction for unknown confounding
factors that may affect formation of heterotopic bone.
Secondly, although our population was homogeneous
with respect to gender, aetiolo gy, type of prosthesis, sur-
gical approach, and rehabilitation protocol, the index
group was significan tly older . Adva nced age at surg ery is
a risk factor for HO after THA [15]. This would suggest
that the effect of pulsed lavage is even greater than our
results have shown. Thirdly, we only used a radiological
outcome at one-year follow-up and no clinical scores
were assessed. However, several studies have shown that
radiological Brooker grades 3 and 4 correlate positively
with clinically relevant HO formation [4-6]. Also a fol-
low-up of one year seems appropriate because HO will
be evident and stable in size at 6 weeks postoperatively,
while maturation of formed HO takes place up to 6
months after surgery [16].
A wide range (8% - 90%) in HO incidence after THA
has been described.[1,11,15] These differences may de-
pend on the reliability of the HO classification system
used. Different systems have been suggested for assess-
ing HO formation. Toom et al. [17] reported high in-
ter-observer reliability for three methods assessing HO
formation (Arcq (k = 0.9) [18] DeLee (k = 0.9) [19] and
Della Valla (k = 0.9) [5]). We used the widely accepted
and most frequently used Brooker's classification [20]. A
recent study showed good inter-observer agreement (k =
0.8) for assessing HO formation with the Brooker’s clas-
sification [17]. We also found good inter-observer agree-
ment (k = 0.7) between three experienced blinded ortho-
paedic surgeons for scoring HO on standard anteroposte-
rior pelvic radiographs. More likely, heterogeneity a-
mong studies contributes to the various incidences of HO
reported. A recent report on 134 patients who had a ce-
mentless hydroxyapatite-coated THA without pharma-
cologic or radiotherapeutic prophylaxis, showed 67% HO
formation [21]. While another randomized study reported
31% HO formation in 97 patients treated with unce-
mented total hip arthroplasty [22]. A systematic over-
view of 201 studies showed the incidence of any het-
erotopic bone formation to be 43% after total hip re-
placement, the incidence of severe HO was 9%. Informa-
tion about patient characteristics, use of heterotopic bone
formation prophylaxis, other NSAID use and details of
surgery and follow up were very incompletely reported
and generally available for less than one-quarter of the
included studies [20]. Risk factors for HO have been re-
cognized and include male sex, posterolateral approach
and hypertrophic osteoarthritis [10,11]. Our study popu-
lation was selected for these known risk factors, and HO
formation was found in 51% of the cases in the index
group versus 58% in the control group.
The incidence of clin ically relevant HO was 3% in the
index group versus 17% in the control group, which was
significantly (p = 0.04) different. This suggests a benefi-
cial effect on severe HO formation by using pulsed lav-
age. The exact aetiology of HO formation is still un-
known. Some have suggested that mesenchymal pluripo-
tential stem cell release from bone during surgery may
act as a stimulus [23,24]. McCarthy proposed a chain of
events in which four factors are important: surgical
trauma that leads to haematoma; protein messengers re-
lease from traumatized cells or inflammatory cells that
arrive at the location of trauma; protein messengers acti-
vation of mesenchymal cells to transform into osteob lasts
or chondroblasts, and a proper environmental condition
that enables HO formation [8]. Pulsed lavage may have
the potential to washout mesenchymal pluripotential stem
cells and messenger proteins, or to distort the proper en-
vironmental condition needed for HO formation. Clinical
studies on the effect of pulsed lavage in THA treatment
are rare, and only one study reported on pulsed lavage
and HO formation [1]. This small and heterogeneous
study found no positive effect for pulsed over syringe
lavage, but lacked a true control group which had no
lavage at all. A recent Japanese trial on 1000 uncemented
THA’s through a postero lateral approach reported severe
Copyright © 2011 SciRes. OJO
Pulsed Lavage in Cementless Total Hip Arthroplasty Reduces the Incidence of Brooker
12 Grade 3 and 4 Heterotopic Ossifications
HO formation in only 1% of the cases. A large portion of
patients (40%) received NSAIDs postoperatively, and all
wounds were irrigated with 1000 mL of saline [25]. Al-
though no pulsatile lavage was used wound lavage may
have had an additional beneficial effect, causing the low
incidence of HO.
5. Conclusions
In conclusion our study showed significantly less Brooker
grades 3 and 4 heterotopic ossifications after cementless
THA in male patients treated with pulsed lavage. This
suggests a positive effect of pulsed lavage on preventing
clinically relevant HO formation. Other studies suggested
that lavage in general reduces the incidence of HO after
THA. Future prospective randomized trials should eluci-
date whether pulsed lavage, in addition to syringe lavage
and no lavage at all, is superior.
6. Acknowledgements
We thank I. van de Akker-Scheek, PhD, for her contribu-
tion to the manuscript.
[1] R. J. Sneath, F. D. Bindi, J. Davies and E. J. Parnell, “The
Effect of Pulsed Irrigation on the Incidence of Het-
erotopic Ossification after Total Hip Arthroplasty,” Jour-
nal of Arthroplasty, Vol. 16, No. 5, 2001, pp. 547-551.
[2] A. Toom, T. Haviko and L. Rips, “Heterotopic Ossifica-
tion after Total Hip Arthroplasty,” International Ortho-
paedics, Vol. 24, No. 6, 2001, pp. 323-326.
[3] B. C. Neal, A. Rodgers, T. Clark, H. Gray, I. R. Reid, L.
Dunn, et al., “A Systematic Survey of 13 Randomized
Trials of Non-Steroidal Anti-Inflammatory Drugs for the
Prevention of Heterotopic Bone Formation after Major
Hip Surgery,” Acta Orthopaedica, Vol. 71, No. 2, 2000,
122-128. doi:10.1080/000164700317413076
[4] A. Buvanendran, J. S. Kroin and R. A. Berger, “Preopera-
tive Cyc-Looxygenase-2 Inhibitor Treatment Reduces the
Incidence of Hheterotopic Ossification after Hip Arthro-
plasty: Six-Month Follow-Up,” Anesthesiology, Vol. 107,
No. 2, 2007, 358-359.
[5] A. G. Della-Valle, P. S. Ruzo, V. Pavone, E. Tolo, D. N.
Mintz and E. A. Salvati, “Heterotopic Ossification after
Total Hip Arthro-Plasty: A Critical Analysis of the
Brooker Classification and Proposal of a Simplified Rat-
ing System,” Journal of Arthroplasty, Vol. 17, No. 7,
2002, 870-875.
[6] P. Kjaersgaard-Andersen and M. A. Ritter, “Prevention of
Formation of Heterotopic Bone after Total Hip Arthro-
plasty,” Journal of Bone and Joint Surgery, Vol. 73, No.
6, 1991, 942-947.
[7] O. Kolbl, D. Knelles, T. Barthel, U. Kraus, M. Flentje and
J. Eulert, “Randomized Trial Comparing Early Postopera-
tive Irradiation vs. the Use of Nonsteroidal Antiinflam-
matory Drugs for Prevention of Heterotopic Ossification
Following Prosthetic Total Hip Replacement,” Interna-
tional Journal of Radiation Oncology, Biology and Phys-
ics, Vol. 39, No. 5, 1997, pp. 961-966.
[8] E. F. McCarthy and M. Sundaram, “Heterotopic Ossifica-
tion: A Review,” Skeletal Radiology, Vol. 34, No. 10,
2005, pp. 609-619. doi:10.1007/s00256-005-0958-z
[9] T. C. Lo and W. L. Healy, “Re-Irradiation for Prophylaxis
of Heterotopic Ossification after Hip Surgery,” British
Journal of Radiology, Vol. 74, No. 882, 2001, pp. 503-
[10] B. R. Horwitz, N. L. Rockowitz, S. R. Goll, R. E. Booth,
Jr., R. A. Balderston, R. H. Rothman, et al., “A Prospec-
tive Randomized Comparison of Two Surgical Ap-
proaches to Total Hip Arthroplasty,” Clinical Orthopae-
dics and Related Research, Vol. 291, 1993, pp. 154-163.
[11] J. G. Nollen and F. Q. van Douveren, “Ectopic Ossifica-
tion in Hip Arthroplasty. A Retrospective Study of Pre-
disposing Factors in 637 Cases,” Acta Orthopaedica, Vol.
64, No. 2, 1993, pp. 185-187.
[12] K. Knahr, M. Salzer, R. Eyb, P. Frank and W. Blauen-
steiner, “Heterotopic Ossification with Total Hip Endo-
prostheses in Various Models of Thrombosis Prophy-
laxis,” Journal of Arthroplasty, Vol. 3, No. 1, 1988, pp.
1-8. doi:10.1016/S0883-5403(88)80047-0
[13] D. C. Ayers, C. M. Evarts and J. R. Parkinson, “The Pre-
vention of Heterotopic Ossification in High-Risk Patients
by Low-Dose Radiation Therapy after Total Hip Arthro-
plasty,” Journal of Bone and Joint Surgery, Vol. 68, No.
9, 1986, pp. 1423-1430.
[14] A. F. Brooker, J. W. Bowerman, R. A. Robinson, L. H.
Riley Jr., “Ectopic Ossification Following Total Hip Re-
placement. Incidence and a Method of Classification,”
Journal of Bone and Joint Surgery, Vol. 55, No. 8, 1973,
pp. 1629-1632.
[15] P. Kjaersgaard-Andersen, J. Sletgard, C. Gjerloff and F.
Lund, “Heterotopic Bone Formation after Noncemented
Total Hip Arthroplasty. Location of Ectopic Bone and the
Influence of Postoperative Antiinflammatory Treatment,”
Clinical Orthopaedics and Related Research, Vol. 252,
1990, pp. 156-162.
[16] M. A. Ritter and R. B. Vaughan, “Ectopic Ossification
after Total Hip Arthroplasty. Predisposing Factors, Fre-
quency, and Effect on Results,” Journal of Bone and
Joint Surgery, Vol. 59, No. 3, 1977, pp. 345-351.
[17] A. Toom, K. Fischer, A. Martson, L. Rips and T. Haviko,
“Inter-Observer Reliability in the Assessment of Het-
erotopic Ossification: Proposal of a Combined Classifica-
tion,” International Orthopaedics, Vol. 29, No. 3, 2005,
pp. 156-159. doi:10.1007/s00264-004-0603-9
[18] M. Arcq, “Die Paraartikulären Ossifikationeneine Kom-
Copyright © 2011 SciRes. OJO
Pulsed Lavage in Cementless Total Hip Arthroplasty Reduces the Incidence of Brooker
Grade 3 and 4 Heterotopic Ossifications
Copyright © 2011 SciRes. OJO
plikation der Totalhüftendoprothese des Hüftgelenks,” Arch
Orthop Unfallchir, Vol. 77, 1973, pp. 108-131.
[19] J. DeLee, A. Ferrari and J. Charnley, “Ectopic Bone For-
mation Following Low Friction Arthroplasty of the Hip,”
Clinical Orthopaedics and Related Research, Vol. 121,
1976, pp. 53-59.
[20] B. Neal, H. Gray, S. MacMahon and L. Dunn, “Incidence
of Heterotopic Bone Formation after Major Hip Surgery,”
ANZ Journal of Surgery, Vol. 72, No. 11, 2002, 808-821.
[21] R. J. Kasetti, A. A. Shetty and C. Rand, “Heterotopic Os-
sification after Uncemented Hydroxyapatite-Coated Pri-
mary Total Hip Arthroplasty,” Journal of Arthroplasty,
Vol. 16, No. 8, 2001, pp. 1038-1042.
[22] K. R. Rama, P. A. Ve ndittoli, M. Ganapat hi, R. Borgmann ,
A. Roy and M. Lavigne, “Heterotopic Ossification after
Surface Replacement Arthroplasty and Total Hip Arthro-
plasty: A Randomized Study,” Journal of Arthroplasty,
Vol. 24, No. 2, 2009, pp. 256-262.
[23] J. E. Puzas, C. M. Evarts and J. S. Brand, “The Stimulus
for Bone Formation,” Hip, 1987, pp. 25-38.
[24] B. J. Thomas, “Heterotopic Bone Formation after Total
Hip Arthroplasty,” Orthopedic Clinics of North America,
Vol. 23, No. 2, 1992, pp. 347-358.
[25] T. Higo, M. Mawatari, M. Shigematsu and T. Hotokebu-
chi, “The Incidence of Heterotopic Ossification after Ce-
mentless Total Hip Arthroplasty,” Journal of Arthroplasty,
Vol. 21, No. 6, 2006, pp. 852-856.