International Orthopaedics (SICOT)
DOI 10.1007/s00264-015-2726-6
ORIGINAL PAPER
Randomized controlled trial of conventional versus modern
surgical dressings following primary total hip and knee
replacement
Jean Langlois & Amine Zaoui & Camille Ozil &
Jean-Pierre Courpied & Philippe Anract &
Moussa Hamadouche
Received: 11 February 2015 / Accepted: 25 February 2015
# SICOT aisbl 2015
Abstract
Purpose We prospectively compared two types of dressing
(conventional gauze-based versus absorbing hydrofibre) after
primary total hip (THA) or knee (TKA) arthroplasties.
Methods Eighty candidates for THA (n=40) or TKA (s=40)
were randomized: gauze-based versus hydrofibre absorbing
(Aquacel®, ConvaTec). The two groups were comparable at
baseline.
Results There was a statistically significant decrease of dress-
ing changes in the hydrofibre group (p=0.0006). Two patients
from the conventional group presented minor wound compli-
cations. Nurses’ satisfaction was significantly higher in the
hydrofibre group considering the adherence (p=0.04) and
flexibility (p=0.03). Patients experienced a higher satisfaction
with respect to ease of movement (p=0.01) in the hydrofibre
group. The cosmetic appearance of the scars six weeks after
surgery was found to be similar between groups.
Conclusions Our findings support an overall improved com-
fort for the patients and the medical staff by using hydrofibre
dressings after primary THA and TKA. The reduction of re-
quired dressing changes was observed also.
Keywords Wound dressing . Hip arthroplasty . Knee
arthroplasty . Patient satisfaction . Nurse satisfaction
J. Langlois (*) : A. Zaoui : J.<P. Courpied : P. Anract :
M. Hamadouche
Department of Orthopaedic and Trauma Surgery, Hôpital Cochin,
Université Paris-Descartes, Assistance Publique-Hôpitaux de Paris,
27 Rue du Faubourg St Jacques, 75014 Paris, France
e-mail: jeangast@gmail.com
C. Ozil
Department of Reconstructive Surgery, Hôpital Européen Georges
Pompidou, Université Paris-Descartes, Assistance
Publique-Hôpitaux de Paris, 20 Rue Leblanc, 75015 Paris, France
Introduction
The increasing numbers of lower limb arthroplasties, com-
bined with a considerable pressure on healthcare providers
to increase patients’ satisfaction and minimize overall costs,
have instigated new clinical pathways and multidisciplinary
protocols [1, 2]. Promoting optimal wound healing and
preventing surgical site infection is one of the cornerstones
of these regimens. Wound healing involves both local and
systemic host factors [3], surgical considerations [4], but also
relies on the dressing that is applied. An ideal environment
promoting healing should be moist, warm and clean [5].
Therefore, a perfect dressing should absorb sufficient exudate,
while being permeable enough to control a moist environ-
ment. In addition, a dressing must act as a watertight barrier
against the external environment, while also allowing the
wound to breathe [6].
Aside from questions related to the moisture-permeability
balance, total joint arthroplasty wounds present morphologic
specificities [7]. Dressings over joints should accommodate
motion allowing wide and free range of movements with no
risk of friction. Moreover, following joint replacement, the
skin is often frail and subject to postoperative oedema fluctua-
tions. Consequently an ideal dressing devoted to the
arthroplasty field should allow wound inspection. If not trans-
parent, this innovative dressing should therefore come with
sufficient warranties to allow a safe reduction of the number
of changes [7]. In addition, changing dressings remains often a
painful procedure [8], which is time- and cost-consuming, and
still represents a contamination risk.
The hydrofibre (Aquacel®, ConvaTec, Greensboro, NC,
USA) dressing is a highly absorbent sterile hydrocolloid
(carboxymethylcellulose) wound dressing that converts into
soft gel when it comes in contact with wound exudates. Its

main advantage is based on the concept that the dressing can
be left without regular changes, while providing a micro-
environment that is supposed to facilitate healing. Based on
all these considerations, we conducted a randomized con-
trolled trial (RCT) to compare the modalities of use and the
effect of postoperative surgical dressings (conventional versus
hydrofibre) on patients and nurses’ satisfaction in two pro-
spective randomized series of primary total hip or knee
arthroplasty. We hypothesized that the use of hydrofibre dress-
ings would reduce the number of dressing changes during
hospital stay.
Materials and methods
Study design
A power analysis, based on the expected difference of num-
ber of dressing changes, was performed. The number of
patients to include was calculated assuming that a mean
number of two (SD 1) dressing changes during hospital stay
would be necessary in the conventional dressing group ver-
sus one in the hydrofibre dressing group. With a two-sided
risk alpha of 2.5 % and a power (1 - beta) of 90 %, calcu-
lation indicated that 25 patients would be needed in each
group. After adjustment for non-compliance/cross-over, we
decided to include 40 patients in each arm. The patients
were randomized using a computer-generated block ran-
domization scheme to have either conventional or
hydrofibre dressing. The randomization was based on the
order of patient presentation, so each patient was random-
ized individually regardless of severity of osteoarthritis and
co-morbid situation. To avoid surgeon bias, randomization
was stratified by surgeon. They all had given informed con-
sent, and the study was IRB approved.
Patients
This RCT was conducted over a six-month period in a sin-
gle institution. Eighty consecutive patients with unilateral
primary degenerative osteoarthritis of the hip (n =40) or
knee (n=40) requiring total joint replacement were enrolled
in the study. The inclusion criteria were patients aged be-
tween 18 and 95 years and able to understand information.
The exclusion criteria were prior operative local procedure
around the joint, past local infection, and advanced cancer.
Demographic characteristics including age, gender, body
mass index (BMI), smoking status, medical comorbidities
(diabetes mellitus, renal insufficiency and skin or vascular
disorders), in addition to medications (steroids) were
recorded.
International Orthopaedics (SICOT)
Procedures
All surgical procedures were carried out by three experi-
enced lower limb reconstruction surgeons. Total hip re-
placements were performed using either an anterior ap-
proach in supine position as modified by Röttinger [9], or
a transtrochanteric approach, with the patient in lateral po-
sition. Total knee replacements were performed through a
standard medial parapatellar approach. A tourniquet was
used according to the surgeons’ usual practice (unless there
was contra-indication related to peripheral arterial disease),
whereby two of the surgeons always used tourniquet and
one did not use it. Surgical closure was standardized in-
volving three layers. The deep layer was repaired using a
resorbable number 0 Vicryl® (Ethicon, Johnson & Johnson,
Norderstedt, Germany). A subdermal layer was closed with
a 2–0 Vicryl®, followed by simple interrupted stitches for
skin closure with 3–0 monofilament nylon. The suction
drain exits were carefully positioned away from the wound,
so their removal would not disturb wound dressing. In the
hydrofibre group, an Aquacel® dressing pad was placed in
intimate contact with the wound. It was not covered by any
extra dressing in order to facilitate direct inspection. In the
conventional group, sterile gauzes were spread over the
wound and held in place with a crepe bandage. Wound
dressings applied intra-operatively were secondarily
changed under a specific protocol for each group. In the
conventional dressing group, the dressing change was
scheduled between day one and day three postoperatively,
associated to a second change on the day of discharge.
Gauzes were then replaced by a conventional adhesive
pad (Mepore, Mölnlycke Health Care, Göteborg, Sweden).
In the hydrofibre absorbing group, the change was only
scheduled the day of discharge. In both groups, an extra
change of dressing was performed in case of saturation with
leakage, major loss of adherence, bleeding, or suspected in-
fection. The post-operative regimen included the administra-
tion of systemic antibiotics for 48 hours, thromboprophylaxis
with low molecular weight heparin (enoxaparin 40 SC/day)
for six weeks, and anti-inflammatory medication (ketoprofen,
100 mg/day for five days) to prevent heterotopic ossification
in the THA group. Immediately after the operation, passive
exercises were begun with the assistance of a physiotherapist,
and were continued until active movement of the hip was
possible. The patients were free to walk after three days with
crutches. Patients were usually discharged five to seven days
postoperatively.
Outcomes
The primary outcome was the effective number of dressing
changes during the hospitalization. Several secondary out-
comes were recorded: (1) wound complications, including
International Orthopaedics (SICOT)
erythema, blistering, delayed healing, maceration, scarring,
necrosis and infection, (2) patients and nurses satisfactions,
and (3) scar cosmetic appearance evaluated six weeks after
surgery by a plastic surgeon that was blinded to the dressing
used and not involved in the surgical procedures. Satisfaction
was recorded the day of discharge using self-administered
questionnaires. For patients, parameters evaluated included
pain during dressing change, anxiety, pain outside change,
ease of movement, itching/stinging/burning/odour, and over-
all satisfaction. For nurses, parameters were ease of use, pa-
tient apprehension, patient pain, adherence, bleeding,
conformability, and overall satisfaction. Likert scale discrete
[10] answers were assigned numbers for the analysis (0: high-
ly unsatisfied; 1: not satisfied; 2: fairly satisfied; 3: satisfied; 4:
highly satisfied). Wound cosmetic aspect was evaluated using
three different validated scales: Stony Brook [11], visual ana-
log cosmetic [12], and categorical (poor, acceptable or excel-
lent categories).
Statistical analysis
Statistical analyses were performed with Real Statistics Re-
source Pack software, version 2.10 (Miller Place, NY, USA).
Normal distribution was tested by the Shapiro–Wilk test. If the
distribution was normal, the parametric Student’s t-test was
used for quantitative variables and chi-squared test for quali-
tative data. Otherwise, nonparametric Mann–Whitney test and
Fischer’s exact tests were used. A p value of less than 0.05 was
considered to be significant.
Assessed for eligibility (n=85)
Enrolled (n=80)
THA (n=40)
20 patients randomized 20 patients randomized
to Hydrofiber to Conventional
6 weeks follow-up 6 weeks follow-up
- Lost to F/U (n=0) - Lost to F/U (n=0)
- Questionnaire (n=19) - Questionnaire (n=18)
- Scar evaluation - Scar evaluation
Fig. 1 Study protocol
Results
Of the 80 included patients, a total of 75 satisfaction evaluation
forms (93.8 %) were obtained and 73 wound scars (91.3 %) were
evaluated with no patient lost to follow-up (Fig. 1). There was no
significant difference between the conventional and hydrofibre
groups in terms of demographics (Table 1). The mean number
of dressing changes in the conventional group was two (ranges,
two to five) versus one (ranges, one to four) in the hydrofibre
group (p=0.0006) (Table 2). Two patients from the conventional
dressing group underwent a minor wound complication (blistering
in one case, erythema without proven infection in the other case),
that did not require any additional surgical procedure nor antibiotic
therapy. No allergic-like reaction was recorded in the hydrofiber
group. Nurses’ satisfaction was significantly higher in the
hydrofibre group considering the adherence (p=0.04) and the
flexibility (p=0.03) categories. Patients in the hydrofibre group
experienced a higher satisfaction considering their ease for move-
ment (p=0.01). Other criteria (apprehension, pain, bleeding during
change, odour or itching) were not found to be significantly dif-
ferent between groups with the numbers available (Table 2). Six
weeks after surgery, the overall wound scar aspect was found to be
similar between groups, according to the three scales (Table 2).
Discussion
Aside from the question of suction drainage [13], the identifi-
cation of an optimal dressing for total joint arthroplasty
Excluded (n=5)
- Not meeting inclusion criteria (n=5)
- Declined to participate (n=0)
TKA (n=40)
20 patients randomized 20 patients randomized
to Hydrofiber to Conventional
6 weeks follow-up 6 weeks follow-up
- Lost to F/U (n=0) - Lost to F/U (n=0)
- Questionnaire (n=19) - Questionnaire (n=19)
- Scar evaluation - Scar evaluation
 International Orthopaedics (SICOT)

Table 1 Demographic characteristics of the 80 included patients dressing are numerous: (1) an optimal moist balance for
Dressing Conventional Hydrofiber p-value healing, sequestering fluids to avoid maceration while releas-
(n=40) (n=40) ing a gel that maintains a humid environment, (2) its imper-
meable nature also strongly impacts the interval between
Age (year) a 68.3±10.3 71.1±11.4 0.34 dressing changes: despite being highly dependent on the
Sex ratio (female/male) b 29 F / 11 M 27 F / 13 M 0.63 amount of exudate produced by the wound, a hydrofibre
BMI (kg/m²) a 26.8±4.4 25.4±4.3 0.22 dressing may frequently be of help in place up to seven days.
Diabetes mellitus b 6 2 0.26 In the current RCT following total joint arthroplasty proce-
Steroid dependance b 4 6 0.74 dures, the use of hydrofibre dressings significantly decreased
Severe renal insufficiency b 0 0 1 the number of dressing changes, and leads to higher levels of
Smoking b 2 7 0.15 satisfaction from both the nurses’ and patients’ perspectives,
Skin or vascular disorder b 1 1 1 considering respectively adherence and flexibility in one side,
and ease of movement on the other side. We did not show any
Data are expressed as means ± standard deviations, or absolute number of
cases (n) statistical difference in terms of peri-wound injury or short-
M male, F female, BMI body mass index
term cosmetic appearance of scars.
a
Student’s t-test b Two-tailed Fisher’s exact test
Our study has several limitations. We recognized having
controlled most but not every wound healing confounding
wounds is an area sometimes neglected by orthopaedic sur- factor, such as protein or vitamin C deficiencies, liver or thy-
geons, although wound complications can lead to serious con- roid diseases [14]. Our protocol did not register the compara-
ditions. In this context, potentials advantages of the hydrofibre tive time to complete wound healing, and was not powered
enough to compare wound complications, as the natural oc-
currence of such adverse events is very low. We also acknowl-
Table 2 Comparative dressing performance edge that the potential care improvement related to the dress-
ing would not be sufficient by itself to demonstrate a major
Dressing Conventional Hydrofiber p-value
clinical difference with a reasonable inclusion number of pa-
Number of dressing changes a 2±0.94 1±0.91 0.0006 tients in only one centre. Further trials will be needed to eval-
Peri-wound adverse reaction b 2 0 0.49 uate the potential benefits from these hydrofibre dressings
Nurses satisfaction c both in terms of prevention of wound infections or reduction
Ease of use a 3±0.59 4±0.49 0.13
of length of stay.
Patient apprehension a 3±0.73 4±0.59 0.38
Published data regarding the clinical performance of
Patient pain a 4±0.69 4±0.66 0.44
hydrofibre dressings are still limited and conflicting. The Na-
tional Institute for Health and Clinical Excellence (NICE)
Adherence a 3±1.09 4±0.89 0.04
guidelines suggest that no clinical evidence exists to support
Bleeding a 4±1.07 4±0.56 0.21
one dressing over another [15]. This opinion has been sup-
Conformability a 3±0.71 4±0.49 0.03
ported recently by Collins et al. [4] in a recent exhaustive
Overall experience a 3±0.64 4±0.67 0.26
review. In addition, Ubbink et al. [16] demonstrated that the
Patients satisfaction c
new occlusive dressings provided no advantage in terms of
Pain during change a 4±0.60 4±0.48 0.54
wound healing. On the other side, another body of the litera-
Anxiety a 4±0.75 4±0.75 0.65
ture indicates that these modern absorbing dressings might be
Pain outside change a 3±0.90 3±0.97 0.20
associated with shorter wear time, less changes required, and
Ease of movement a 3±0.81 4±0.85 0.02
less blistering [17–19], together with reduced surgical site
Itching/stinging/burning/odour a 4±1.05 4±0.80 0.62
infection [14]. However these studies were of low level of
Overall experience a 3±0.51 3±0.80 0.97
evidence, being either retrospective, or non-randomized.
Cosmesis six weeks postop
Moreover, patients’ and nurses’ satisfactions were not taken
Stony Brook scale a 0±1.62 1±1.71 0.55
into account, neither was the cosmetic appearance of the scar.
Visual analog cosmetic scale a 50±19.3 47.5±15.4 0.53
In the absence of solid scientific proof, the selection of
Categorical scale a 0±0.71 0.5±0.63 0.66 postoperative dressing should in fact be based on a compro-
Data are expressed as median ± standard deviations mise between evidence-based and patient-focused consider-
a
Mann–Whitney test ations. Satisfaction questionnaires evaluating patient comfort
b
Fisher exact test
are indeed of major importance [20]. Studies incorporating
c
Likert scale discrete answers were assigned numbers for the analysis (0:
specifically such outcomes are yet scarce [21, 22]. In the as-
highly unsatisfied; 1: not satisfied; 2: fairly satisfied; 3: satisfied; 4: highly sessment of a dressing an additional difficulty may be attrib-
satisfied) uted to patients’ difficulty to differentiate postoperative
International Orthopaedics (SICOT)
wound pain from discomfort dressing-related. In addition,
pain is difficult to measure objectively. However, the use of
dedicated tools combined with thorough explanations can pro-
vide reliable results [23]. As demonstrated by Hamilton et al.
[24], the patient’s hospital experience is one of the three fac-
tors that determine its overall satisfaction, emphasizing that
Bhow we deliver healthcare^ might be as important, if not
more, than Bthe specifics of what we deliver^.
This level of evidence I study is one of the first to compare
two types of dressings after THA and TKA considering vari-
ous outcomes. To achieve best practice, our goal was to act on
research findings in a cohesive (including nurses’ opinion)
and patient centred way (including patient’s opinion).
Hydrofibre dressings should be considered with regards to
the management of wounds after total joint arthroplasty of
the lower limb. For at least equal healing performance, the
benefits in terms of patient comfort and nurses’ spared time
are evident. Larger well-designed trials should confirm their
definitive cost-effectiveness.
Conflict of interest None.
References
1. Mertes SC, Raut S, Khanduja V (2013) Integrated care pathways in
lower-limb arthroplasty: are they effective in reducing length of hos-
pital stay? Int Orthop 37(6):1157–1163
2. Ibrahim MS, Khan MA, Nizam I, Haddad FS (2013) Peri-operative
interventions producing better functional outcomes and enhanced
recovery following total hip and knee arthroplasty: an evidence-
based review. BMC Med 11:37
3. Guo S, Dipietro LA (2010) Factors affecting wound healing. J Dent
Res 89(3):219–229
4. Collins A (2011) Does the postoperative dressing regime affect
wound healing after hip or knee arthroplasty? J Wound Care 20(1):
11–16
5. Benbow M (2008) Exploring the concept of moist wound healing and
its application in practice. Br J Nurs 17(15):S4–S8
6. Tustanowski J (2009) Effect of dressing choice on outcomes after hip
and knee arthroplasty: a literature review. J Wound Care 18(11):449–
458
7. Cosker T, Elsayed S, Gupta S, Mendonca AD, Tayton KJ (2005)
Choice of dressing has a major impact on blistering and healing
outcomes in orthopaedic patients. J Wound Care 14(1):27–29
8. Whitehouse JD, Friedman ND, Kirkland KB, Richardson WJ, Sexton
DJ (2002) The impact of surgical-site infections following orthopae-
dic surgery at a community hospital and a university hospital: adverse
quality of life, excess length of stay, and extra cost. Infect Control
Hosp Epidemiol 23(4):183–189
9. Bertin KC, Röttinger H (2004) Anterolateral mini-incision hip re-
placement surgery: a modified Watson-Jones approach. Clin Orthop
Relat Res 429:248–255
10. Jamieson S (2004) Likert scales: how to (ab)use them. Med Educ
38(12):1217–1218
11. Singer AJ, Arora B, Dagum A, Valentine S, Hollander JE (2007)
Development and validation of a novel scar evaluation scale. Plast
Reconstr Surg 120(7):1892–1897
12. Quinn JV, Drzewiecki AE, Steill IG, Elmslie TJ (1995) Appearance
scales to measure the cosmetic outcomes of healed lacerations. Am J
Emerg Med 13:229–231
13. Zhou XD, Li J, Xiong Y, Jiang LF, Li WJ, Wu LD (2013) Do we
really need closed-suction drainage in total hip arthroplasty? A meta-
analysis. Int Orthop 37(11):2109–2118
14. Cai J, Karam JA, Parvizi J, Smith EB, Sharkey PF (2013) Aquacel
surgical dressing reduces the rate of acute PJI following total joint
arthroplasty: a case–control study. J Arthroplasty 29(6):1098–1100
15. National Institute for Health and Clinical Excellence (NICE) (2008)
Surgical site infection: Prevention and treatment of surgical site in-
fection. NICE. http://publications.nice.org.uk/surgical-site-infection-
cg74. Accessed 06 March 2015
16. Ubbink DT, Vermeulen H, Goossens A, Kelner RB, Schreuder SM,
Lubbers MJ (2008) Occlusive vs gauze dressings for local wound
care in surgical patients: a randomized clinical trial. Arch Surg
143(10):950–955
17. Abuzakuk TM, Coward P, Shenava Y, Kumar VS, Skinner JA (2006)
The management of wounds following primary lower limb
arthroplasty: a prospective, randomised study comparing hydrofibre
and central pad dressings. Int Wound J 3(2):133–137
18. Clarke JV, Deakin AH, Dillon JM, Emmerson S, Kinninmonth AW
(2009) A prospective clinical audit of a new dressing design for lower
limb arthroplasty wounds. J Wound Care 18(1):5–8
19. Hopper GP, Deakin AH, Crane EO, Clarke JV (2012) Enhancing
patient recovery following lower limb arthroplasty with a modern
wound dressing: a prospective, comparative audit. J Wound Care
21(4):200–203
20. Pacault-Legendre V, Courpied JP (1999) Survey of patient satisfac-
tion after total arthroplasty of the hip. Int Orthop 23(1):23–30
21. Robertsson O, Dunbar M, Pehrsson T, Knutson K, Lidgren L (2000)
Patient satisfaction after knee arthroplasty: a report on 27,372 knees
operated on between 1981 and 1995 in Sweden. Acta Orthop Scand
71(3):262–267
22. Scott CE, Bugler KE, Clement ND, MacDonald D, Howie CR, Biant
LC (2012) Patient expectations of arthroplasty of the hip and knee. J
Bone Joint Surg (Br) 94(7):974–981
23. Mahomed N, Gandhi R, Daltroy L, Katz JN (2011) The self-
administered patient satisfaction scale for primary hip and knee
arthroplasty. Arthritis 591253:1–6
24. Hamilton DF, Lane JV, Gaston P, Patton JT, Macdonald D, Simpson
AH, Howie CR (2013) What determines patient satisfaction with
surgery? A prospective cohort study of 4709 patients following total
joint replacement. BMJ Open 3(4)