Wound complications are commonly acknowledged risks in primary total knee and total hip arthroplasty. Additional surgery is occasionally required to manage these adverse events. In total knee procedures, the proximal half of the skin incision is subjected to significant stretch, which can lead to wound separation and abnormal scarring. With total hip procedures using a direct anterior approach, the proximal portion of the incision near the abdominal fold is often problematic, having the potential for wound separation and drainage. Wound additives have been used to address these concerns; however, they may or may not improve wound healing in the primary surgical site and remain controversial regarding their effectiveness and expense. In the current health care environment, any additional treatment costs must lead to significant benefit to be incurred.
Platelet-rich plasma (PRP) has become a popular additive that may promote wound healing. It is derived from the patient's autologous blood as a source of concentrated platelets containing growth factors and cytokines. In a recent meta-analysis review, Sclafani and Azzi1 revealed that most of the animal and human clinical studies have reported favorable results in wound healing. Within the animal model, collagen deposition and epithelialization were enhanced with PRP.2 However, human results using PRP in knee replacements have been mixed. Morishita et al3 failed to find improved outcomes in total knee replacements with the use of PRP. Everts et al4 showed improved range of motion following the application of platelet gel in total knees. Aggarwal et al5 documented a decrease in both postoperative drainage volume and blood loss after PRP use in knee replacements. On the basis of these studies, PRP has the potential to benefit outcomes in joint replacement arthroplasty and to reduce postoperative wound complications.
CellerateRX (Wound Management Technologies, Inc, Fort Worth, Texas) powder (activated collagen) was recently introduced to assist with wound healing because of its ability to expand 30 times its volume when placed in a moist environment. The additive's wound healing potential has been revealed in many types of open wounds, and its mechanism of action is thought to be that of a chemoattractant for epithelialization.6 Modulation of cytokine production, with changes in interleukin-6 and tumor necrosis factor, has also been seen with activated collagen.7 Clinically, this activated collagen has been used in association with antibiotics for sepsis originating from open fractures, ulcers, and total joints.8 The use of activated collagen may have the potential to improve wound healing, preventing possible wound separation and the development of seromas, hematomas, and superficial wound infections that occur with joint replacement surgery.
The objective of this study was to determine whether additives would improve wound outcomes and prevent postoperative complications in primary total knee and total hip surgeries. The authors' hypothesis was that incisional separation and final epithelialization would be enhanced by using either PRP or activated collagen. In addition, the authors examined infection rate, blood loss, and reoperation rate in all cohort groups.
Materials and Methods
This study received institutional review board approval as a prospective, randomized, single-blind, controlled study. Written consent to participate was obtained from the first 90 eligible patients with osteoarthritis who were undergoing elective primary total knee or total hip arthroplasty. Patients were excluded if they had inflammatory arthritis, total joint revisions, and known active intravascular clotting disorders. Prior to surgery, the participants were randomized, via a computer-generated algorithm, into 3 groups of 30 each. Group 1 included 30 patients receiving PRP prior to wound closure, group 2 included 30 patients receiving activated collagen prior to wound closure, and group 3 included 30 patients who acted as controls (standard wound closure without additive). Sex, body mass index, comorbidities, American Society of Anesthesiologists (ASA) score, length of hospital stay, and distribution of knee/hip replacements were similar in the 3 groups (Table 1).
All 90 surgeries were performed by a single fellowship-trained arthroplasty surgeon (B.G.E.). The knee surgeries were completed with a cemented pros-thesis (Vanguard; Zimmer Biomet, Warsaw, Indiana), and the hip surgeries were performed with a cementless prosthesis (Taperloc Stem; Zimmer Biomet). Platelet-rich plasma preparation for group 1 patients was performed preoperatively by drawing 60 mL from the antecubital vein. The blood was then processed using the SmartPrep 2 System (Harvest Terumo, Plymouth, Massachusetts) in accordance with the manufacturer's protocol. The PRP was delivered via liquid applicator. Two syringes were used for injection purposes. The first, labeled “platelet rich,” contained 10 mL of PRP and 2.5 mL of calcium chloride (1000 U/mL). The second, labeled “platelet poor,” contained 15 to 25 mL of platelet-poor plasma (PPP) and 2.5 mL of calcium chloride (1000 U/mL). Activated collagen was provided in 5-g bottles by the hospital and administered to the group 2 patients. The surgical wounds were closed in identical fashion. All total knee arthroplasty closures were performed using quad-sparing, minimally invasive surgery with deep fascia closure using 2 Quill (Surgical Specialties Corporation, Wyomissing, Pennsylvania), subcutaneous space closure with 0 Quill, subcuticular closure with 2-0 Quill, and skin closure with the DERMABOND PRINEO Skin Closure System (Ethicon, Somerville, New Jersey). The direct anterior approach using a Hana (Mizuho/OSI, Union City, California) table was completed for all primary total hip arthroplasties with tensor fascia and subcutaneous tissue closure using 0 Quill, subcuticular closure with 2-0 Quill, and skin closure with the DERMABOND PRINEO Skin Closure System. In group 1, 50% of both the PRP and the PPP were injected below the deep fascia. The remaining 50% of the PRP and the PPP were injected throughout the subcutaneous space prior to final skin closure. In group 2, 50% of the activated collagen was placed deep to the fascial layer closure and 50% was placed in the subcutaneous space prior to skin closure. Group 3, acting as a control, had no additives placed in the soft tissues.
A 325-mg dose of aspirin was administered twice daily for 3 weeks for deep venous thrombosis prophylaxis except in patients who were restarted on their preoperative anticoagulation regimen after surgery. Spinal anesthesia was completed for 83% of all surgeries, with the remainder being performed under general anesthesia. Prior to skin incision, all 90 patients received 1 g of tranexamic acid. A periarticular injection that included 200 mg of ropivacaine, 30 mg of ketorolac, 0.1 mg of clonidine, 0.5 mg of epinephrine, and 35 mL of normal saline was infiltrated throughout every surgical layer. Hemovac drains (1/8 inch) were placed in all procedures to record postoperative drainage for the first 15 hours. A preoperative hematocrit and a 48-hour postoperative hematocrit were recorded for each patient. The patients were provided postoperative wound instructions and advised to keep a dry dressing on the incision until the fifth postoperative day, at which time showers were allowed. Because of the DERMABOND seal, the wound was left uncovered after 5 days, unless drainage occurred.
Wound analysis was completed by a medical assistant trained in wound care and blinded to all 3 groups. The wound data were recorded at 2 and 6 weeks postoperatively. A medical review of hospital, rehabilitation institution, and office charts was conducted to determine preoperative and postoperative day 2 hematocrit results, post-index surgical procedures, and infections. Antibiotic administration, for any reason other than the normal 24-hour prophylactic antibiotic, was recorded during the entire 6-week trial. Table 2 lists the individual wound observations completed at both the second and the sixth week postoperative evaluations. Total blood loss was determined by calculating the predicted blood volume according to the Nadler et al9 formula based on height, weight, and sex. This value was then multiplied by the percentage difference in hematocrit results:
Wound Observations Completed at 2 and 6 Weeks Postoperatively
For sample size determinations, a power analysis was completed assuming a power of 0.8 and an alpha of 0.05. The required sample size was 105, with 35 in each group. Analysis of variance was used to assess the statistical significance of blood loss among the groups and was followed by a Bonferroni post hoc test for multiple pairwise comparisons. Separate Pearson's chi-square tests were performed to compare multiple groups for many categorical variables. A Poisson test with a Bonferroni adjustment was performed to determine if the number of complications was different among the groups. Statistical significance was set at P<.05, and analyses were performed with SYSTAT version 13.1 software (Systat Software Inc, Chicago, Illinois).
Ninety consecutive patients were scheduled for primary total knee or total hip replacement and randomized into 3 groups of 30 patients each. Data were collected on wound observations and recorded at 2 and 6 weeks postoperatively (Table 2). One patient in the original group of 90 did not receive the full evaluation and was replaced by another patient to complete the 3 cohorts. No statistically significant differences existed among the 3 groups regarding demographics and clinical characteristics, including sex, body mass index, comorbidities, ASA score, length of hospital stay, and distribution of knee/hip replacements (Table 1).
There were no statistically significant differences between the PRP and the activated collagen wound observations during the 2 time periods. At 2 weeks, there were significant differences in wound erythema, swelling, and drainage between the control group and the PRP and the activated collagen groups. However, at 6 weeks, this tendency did not continue, as there were no statistically significant differences in the wound measurements among the 3 cohorts (Table 2). Measurements of erythema, induration, swelling, ecchymosis, drainage, wound separation, tissue necrosis, and elevated scar were equal in all 3 cohorts at 6 weeks. A total of 5 wound separations were noted at 6 weeks, being distributed across the 3 groups as follows: 1 in the PRP group, 1 in the activated collagen group, and 3 in the control group.
Throughout the entire 6-week period, 12 superficial wound infections were recorded for the 90 participants in the study: 3 in the PRP cohort, 2 in the activated collagen cohort, and 7 in the control cohort. No deep prosthetic joint infections occurred in these 90 patients. Forty-two percent of the entire study population was followed in a rehabilitation setting after hospital discharge. Any antibiotics administered for incisional wounds by the rehabilitation attending staff were recorded as superficial infections.
Four reoperations occurred within the total study population (Table 3). One from the activated collagen cohort involved a wound separation that required irrigation and closure. The other 3 occurred in the control group. Two control patients required hematoma evacuation. One control patient had a superficial infection with purulence, which required irrigation and debridement with closure over drains. Deep cultures remained negative for this patient. No reoperations were required in the PRP group.
Wound Reoperations Among the 3 Groups
A comparison of total postoperative complications among the 3 cohorts led to statistically significant findings. This included a combination of superficial infections and reoperations (Table 4). One patient had both a superficial infection and a reoperation. However, because most of the infections cleared with antibiotics alone, these complications were considered separately for statistical analysis.
Comparison of Total Postoperative Complications Among the 3 Groups
Statistically significant differences were found in this study regarding total blood loss at 48 hours postoperatively (Figure). For each patient, height, weight, and sex were used to calculate the predicted blood volume. This was then multiplied by the percentage change in hematocrit value during the first 48 hours to determine the total blood loss.9 In both the PRP and the activated collagen groups, the total blood loss was significantly less than that of the control group. In this study, total Hemovac drain output after surgery could not be used owing to both inconsistencies and inaccuracies of the nursing staff. No blood transfusions were given to the 90 study patients. Unless indicated for medical reasons, the transfusion protocol was reserved for hematocrit levels of less than 22%. Given the lack of transfusion in all 3 groups, no comparison was possible.
Total mean±SD postoperative blood loss. Blood loss was significantly less in the platelet-rich plasma (PRP) and activated collagen groups. *P<.05 vs PRP and activated collagen.
Proper wound healing is important in preventing superficial infections, hematomas, and seromas that can lead to deep prosthetic joint infections. Accelerated skin healing, especially in areas of excessive skin tension, is desirable. Wound additives that add growth factors or provide a collagen scaffold may or may not improve the ability of the surgical incision to heal. To justify the increased costs associated with these additives, there must be clear evidence that they make a significant difference in the final incisional healing. In this study, 2 additives, PRP and activated collagen, were tested in a prospective, randomized controlled trial of primary total joint arthroplasties to determine if they improved wound outcomes up to 6 weeks postoperatively. Three equal cohort groups were compared: group 1 with PRP, group 2 with activated collagen, and group 3 (control) with no wound additive. Platelet-rich plasma has been shown to be effective for wound healing in both animal and human studies.1.2 Studies of activated collagen have reported positive results in open wounds and proposed benefit to surgical wounds.6,8
This study reported 3 important findings. First, as wound additives, both PRP and activated collagen led to significant improvement in erythema, swelling, and drainage at 2 weeks, but the benefit was no longer present at 6 weeks with all cohorts being equal. Second, the total postoperative complication rate was significantly lower in both the PRP and the activated collagen groups. Third, the total blood loss during the first 48 hours was significantly less when either PRP or activated collagen was added to the surgical wound.
Limited blood loss is important in preventing both hematoma formation and anemia in patients with low preoperative blood counts, as it may prevent allogeneic blood transfusion. Aggarwal et al5 previously showed that PRP did prevent blood loss, but in a small sample of only 40 patients. The need for allogeneic blood transfusions may have been absent in the current study's patient population because of the preoperative administration of tranexamic acid.
No reoperations were necessary in the PRP group. One patient in the activated collagen group required additional surgery because of wound dehiscence. Three reoperations were completed in the control group. Incision and drainage were performed on 1 patient, and hematoma evacuation was performed on the other 2 patients (Table 3). Three of the patients who required reoperation had a direct anterior hip replacement. The literature indicates that this approach may be prone to wound problems because of excessive traction from the minimally invasive procedure.10 The current study confirmed a higher percentage of wound complications in total hips compared with total knees. Furthermore, when the demographics were examined for all patients with postoperative wound healing complications, there was a notable difference in the high-risk patients. These individuals had at least 1 of the following characteristics: morbid obesity, diabetes mellitus, malnutrition, chronic liver failure, preoperative anemia, cardiovascular disorders, chronic renal failure, or current tobacco use. The question of whether this population would benefit from wound additives was not definitively answered by the current study because of the small number of high-risk patients. An ideal future investigation would use a similar design, evaluating high-risk cohorts against a control.
Of the 2 additives tested, neither provided a clear advantage over the other. Both significantly reduced the total blood loss and the total postoperative complication rate. Given the blood loss data, patients requiring heavy preoperative anticoagulation because of hypercoagulation tendencies, anemia, or cardiac issues may benefit from either additive.
Platelet-rich plasma requires veni-puncture and additional personnel to prepare the injection at the time of surgery. Activated collagen is provided in sterile packaging and is easy to apply, which saves surgical time. The cost of each additive, including the expense of additional personnel, may be the determining factor when selecting which one to use at surgery, as the 2 proved equally beneficial.
This study had several limitations. A larger sample might have added to the statistical significance, especially with the tendencies found. In addition, patients with inflammatory arthritis or those who had been treated with immunosuppressive drugs were not included. This sub-population may benefit more from wound additives than patients with osteoarthritis do. Furthermore, the research did not statistically address high-risk patients, who made up a higher percentage of the group with complications.
This study did not support the routine use of either PRP or activated collagen in primary total joint replacements. However, given the significantly decreased blood loss and total complication rate, these additives may be considered for high-risk patients when blood loss and wound healing have been problematic, such as for those with high body mass index, diabetes, and anemia.
- Sclafani AP, Azzi J, Platelet preparations for use in facial rejuvenation and wound healing: a critical review of current literature. Aesthetic Plast Surg. 2015; 39(4):495–505. doi:10.1007/s00266-015-0504-x [CrossRef]
- Jee CH, Eom NY, Jang HM, et al. Effect of autologous platelet-rich plasma application on cutaneous wound healing in dogs. J Vet Sci. 2016; 17(1):79–87. doi:10.4142/jvs.2016.17.1.79 [CrossRef]
- Morishita M, Ishida K, Matsumoto T, Kuroda R, Kurosaka M, Tsumura N. Intraoperative platelet-rich plasma does not improve outcomes of total knee arthroplasty. J Arthroplasty. 2014; 29(12):2337–2341. doi:10.1016/j.arth.2014.04.007 [CrossRef]
- Everts PA, Devilee RJ, Brown Mahoney C, et al. Platelet gel and fibrin sealant reduce allogeneic blood transfusions in total knee arthroplasty. Acta Anaesthesiol Scand. 2006; 50(5):593–599. doi:10.1111/j.1399-6576.2006.001005.x [CrossRef]
- Aggarwal AK, Shashikanth VS, Marwaha N. Platelet-rich plasma prevents blood loss and pain and enhances early functional outcome after total knee arthroplasty: a prospective randomised controlled study. Int Orthop. 2014; 38(2):387–395. doi:10.1007/s00264-013-2136-6 [CrossRef]
- Lulllove EJ. A retrospective analysis of the utilization of “hydrolyzed” type I bovine collagen as demonstrated in a wide variety of wound etiologies in multi-centers. Paper presented at: The Symposium of Advanced Wound Care. ; September 27–29, 2013. ; Las Vegas, NV. .
- Pott GB, Beard KS, Regulski M, Shapiro L. Activated collagen accelerates wound repair and modulates cytokine production in whole blood and PBMC. Paper presented at: 2009 International Cytokine Society. ; October 18–21, 2009. ; Lisbon, Portugal. .
- Farlass B. Use of type 1 bovine collagen surgical, hydrolyzed collagen powder in open distal femur fracture. Poster presented at: Wound Care Innovations Meeting. ; September 18, 2015. ; Fort Worth, TX. .
- Nadler SB, Hidalgo JH, Bloch T. Prediction of blood volume in normal human adults. Surgery. 1962; 51(2):224–232.
- Christensen CP, Karthikeyan T, Jacobs CA. Greater prevalence of wound complications requiring reoperation with direct anterior approach total hip arthroplasty. J Arthroplasty. 2014; 29(9):1839–1841. doi:10.1016/j.arth.2014.04.036 [CrossRef]
|Platelet-Rich Plasma (n=30)||Activated Collagen (n=30)||Control (n=30)|
|Sex, male/female, No.||10/20||13/17||13/17|
|Body mass index, average±SD, kg/m2||28.2±6.7||32±8.1||28.3±5.7|
|Preoperative comorbidities, No.|
| Coronary artery disease||3||2||4|
| Autoimmune disease||2||1||2|
| Morbid obesity (WHO), body mass index >40 kg/m2||2||3||1|
|ASA physical status I/II–VI, No.||26/4||25/5||24/6|
|Length of hospital stay, average, d||2.1||2.2||2.1|
Wound Observations Completed at 2 and 6 Weeks Postoperatively
|2 Weeks||6 Weeks|
|PRP (n=30)||Activated Collagen (n=30)||Control (n=30)||PRP (n=30)||Activated Collagen (n=30)||Control (n=30)|
|Erythema, average, mm||5||4||12||1||2||1|
|Induration, average, mm||2||4||4||0||1||1|
|Swelling, average, mm||7||10||22||2||3||3|
|Ecchymosis, average, mm||10||13||17||0||0||1|
|Wound separation >3 mm, No.||0||1||1||1||1||3|
|Tissue necrosis present, No.||2||3||3||0||1||2|
|Elevated scar, No.||0||0||0||2||1||2|
Wound Reoperations Among the 3 Groups
|Group||Total No. of Wound Reoperations and Types of Surgery|
Wound dehiscence repair (THA)|
Hematoma evacuation (THA)
Hematoma evacuation (THA)
I&D superficial infection (TKA)|
Comparison of Total Postoperative Complications Among the 3 Groups
|Group||Total No. of Postoperative Complications (Infections + Reoperations)|
|Platelet-rich plasma||3 (2 THA, 1 TKA)|
|Activated collagen||3 (3 THA)|
|Control||10 (6 THA, 4 TKA)|