Platelet rich plasma (PRP) is a concentrated platelet product that can be produced from whole blood through multiple commercially available systems, resulting in varying levels of platelet concentration.1,2 Platelets play a crucial role in the signaling cascade of normal wound healing. Activated platelets release the contents of their α-granules, resulting in a deposition of powerful growth factors such as platelet derived growth factor, transforming growth factor-β, vascular endothelial growth factor (VEGF), and epidermal growth factor. 3-5 PRP has been used in many different clinical applications, demonstrating the effectiveness and importance of the product for a variety of medical procedures. For example, percutaneous application of PRP to patients with severe lateral epicondylitis, or tennis elbow, resulted in improved elbow function and reduced pain.6 Early maturation of bony fusion was observed when platelet concentrate was used during lumbar spinal fusions.7 Chronic diabetic foot ulcers treated with PRP achieved increased healing rates compared with the control group receiving standard care.8 Studies by Bhanot and Alex9 show decreased formation of hematoma and seroma, decreased postoperative swelling, and improved healing time for plastic surgeries that included PRP in the treatment. Furthermore, during dental surgeries, the use of PRP has improved bone regeneration around implants.10
There are many different systems on the market that produce PRP, with a high variability in the final products' characteristics, including platelet concentration, volume of output, and fibrinogen concentration. The goal of each of the systems is to concentrate platelets in an easily extractable product. Marx11 suggests that the goal of a PRP is to concentrate platelets three to five times over the baseline platelet counts or at a concentration of at least 1,000,000 platelets/μL. Before conducting PRP efficacy studies, or before trying to draw conclusions on the number of platelets required for each surgical application, it is necessary to be able to accurately determine how many platelets, and consequently the concentration of growth factors, that are applied to the surgical site. However, a twofold problem exists when determining the success of the systems to effectively concentrate the platelets. First, each device outputs a differently concentrated product, and, second, the methods to quantify efficiency of concentration do not appear to have been validated.
Automated hematology analyzers are routinely used for complete blood counts and have been validated to count elements of whole blood, including platelets.12 These validations, however, are not sufficient when counting platelets in PRP for several reasons. Marx13 cited an instance where a Coulter Counter counted clumps of platelets each as one single platelet, giving an falsely low platelet count. However, some hematology analyzer systems are equipped to flag counts with suspected platelet clumping, avoiding this problem. In addition, the manufacturers of each of the analyzers determine an upper limit of the linear range of platelets that can be counted, which PRP can often exceed. Furthermore, platelets in PRPs do not stay suspended in solution and settle within seconds. This phenomenon could be caused by the reduced red blood cell (RBC) number. As a result of the decreased hematocrit, PRPs are optically lighter in color and can cause machine errors, such as incomplete aspiration, indicating that the system did not aspirate enough blood to make an accurate count. Given these limitations to counting PRP with hematology analyzers, effort must be made to solve these issues to count platelets accurately. The aim of this study is to validate a platelet counting method for PRPs using an automated hematology analyzer.
MATERIALS AND METHODS
PRP was produced using a commercially available system (GPS II system, Cell Factor Technologies, Warsaw, IN) according to the product insert instructions. Sixty milliliters of whole, anticoagulated blood was inserted into the GPS disposable. The disposable was then centrifuged for 15 minutes at 3,500 rpm. During the centrifugation step, a tuned-density buoy floats in between the RBCs and the buffy coat, which contains the white blood cells and platelets. A second buoy moves from the top of the disposable, separating the buffy coat and 6 mL of plasma from the rest of the blood plasma. After centrifugation, the excess plasma was removed, and the platelets were resuspended in the PRP by vigorously shaking the disposable for 30 seconds. The PRP was transferred to 13 ×100 mm Elkay tubes and placed on an Ames Aliquot Mixer (Model 4651, Ames Company, Elkhart, IN) before counting.
For each of the following studies, PRPs were made with either bovine blood (Lampire Biological Laboratories, Pipersville, PA) or fresh human blood. Bovine blood was used in many of the tests because it is readily available in large quantities. Therefore, many of the studies were performed with bovine blood but then tested with smaller human sample sizes to ensure the tests are adequate for both species. Both blood sources were anticoagulated with a citrate-based anticoagulant. Stored bovine blood was determined to be viable for platelet counts up to 7 days after harvesting (data not shown).
Validation of Platelet Counts on Hematology Analyzer
The Cell-Dyn 3700 (Abbott Labs, Dallas, TX) hematology analyzer was the system selected to be used in this study because it has a high linear platelet range (2,000 × 103 platelets/μL, per the manufacturer) and includes a veterinary package, which is a software package that allows multiple species to be counted. To validate platelet counts on the hematology analyzer, manual counts were compared with the Cell-Dyn 3700 counts. PRP samples for optical manual counts were prepared in the Unopette microcollection system (Becton Dickinson Vacutainer Systems, Franklin Lakes, NJ) to lyse the RBC and then counted with a hemacytometer (Model 1483, Hausser Scientific, Horsham, PA), using an Olympus BX51 transmission-light microscope at 400× magnification. While counting was performed with the Cell-Dyn 3700, the samples were kept on the rocker before sampling to ensure they were completely mixed and the platelets were evenly distributed.
The number of samples required for statistical significance when comparing manual and hematology analyzer counts was determined from a pilot study (data not shown). One PRP sample was manually counted 10 times, and the standard deviation was determined to be 133 × 103 platelets/μL. With use of the sample number equation, n = [(Zα/2σ)/E]2, where E is the maximum error (determined to be ± 100,000 platelets), σ is the standard deviation, and Zα/2 is the Z statistic determined by the level of confidence (α), the sample size was calculated to be 21 for a 95% confidence level. Therefore, an isolated PRP was counted 21 times manually and with the Cell-Dyn 3700 to validate the hematology analyzer for each PRP preparation. This was repeated for three different PRP preparations.
Increase in Linear Range for Hematology Analyzer
The Cell-Dyn 3700 has a manufacturer-validated linear range limit for platelets of 2,000 × 103 platelets/μL. To increase this range, PRPs were counted on the Cell-Dyn 3700 with platelet counts of approximately 4,000 × 103 platelets/μL. This concentration was selected because it was well above the expected PRP concentrations typically seen clinically. PRPs were created with bovine blood using the GPS II. To increase the platelet concentration to approximately 4,000 × 103 platelets/μL, 1 to 2 mL of plasma was removed from the buffy coat fraction before resuspending the platelets. This resulted in the buffy coat being suspended in less than 10% of the initial starting volume, concentrating the platelets well above what is normally expected clinically. Each concentrated PRP was divided into three samples, one left undiluted, one diluted 1:2 times (50% of initial concentration) with phosphate buffered saline (PBS, pH 7.4), and one diluted 1:4 times (25% of initial concentration) with PBS. All samples were then counted in triplicate on the Cell-Dyn 3700.
Validation of Platelet Counts with Citrate Anticoagulated Blood
The Cell-Dyn 3700 is only validated by the manufacturer for counting platelets in blood anticoagulated with EDTA because this is this most common anticoagulant used for clinical laboratory blood work. However, citrate based anticoagulant is the preferred anticoagulant when producing PRP because it increases platelet viability.14 Whole blood from three human volunteers was drawn into EDTA-coated Vacutainer tubes (Becton Dickinson Vacutainer Systems, Franklin Lakes, NJ) and uncoated Vacutainer tubes (Becton Dickinson Vacutainer Systems, Franklin Lakes, NJ) with 0.8 mL Anticoagulant Citrate Dextrose Solution, Solution A (ACD-A, Citra Anticoagulant, Inc., Braintree, MA). The dilution factor for the ACD-A anticoagulated blood was 1.11. Cell counts were compared between the two anticoagulated blood groups (n = 3 per sample). Statistical analysis compared the counts acquired on the Cell-Dyn for the EDTA blood and the ACD-A blood multiplied by the dilution factor.
Resuspension and Sample Preparation of PRP for Accurate Counts
To determine the necessity of resuspension before platelet counts, three bovine PRP samples were analyzed using the Cell-Dyn 3700 immediately after preparation and were then placed on the Ames Aliquot Mixer and counted every 5 minutes thereafter for a total of 20 minutes.
An additional evaluation was performed to determine whether a small aliquot could be aspirated from the GPS II disposable to provide an accurate representation of the entire PRP fraction. Using a pipette, 0.6 mL of the PRP sample was extracted immediately after vigorous shaking, as described above, and counted on the Cell-Dyn 3700. The remaining PRP sample was placed on a rocker and allowed to mix for 15 minutes and then counted again to compare with the 0.6 mL extraction.
Platelet Counts for PRP Created by GPS II
As a culmination of the validation procedures and platelet concentrator studies, 153 bovine PRP samples, using blood from approximately eight different cows, were created using the GPS II concentration system. The volume of each PRP sample was recorded and each sample was counted for platelets using the Cell-Dyn 3700 three times. The corresponding baseline whole blood counts were also generated three times. Fold increase (PRP platelet concentration/baseline whole blood platelet concentration) and percent recovery [(PRP concentration × volume of PRP)/(baseline blood concentration × 60 mL)] × 100 were calculated. To compare the performance of the GPS II in concentrating human blood as compared with bovine blood, efficacy testing of the GPS II for concentrating human blood was also performed using three human samples.
Data is presented as mean ± one standard deviation. Significant equivalence was determined either with a one-way analysis of variance and a Student-Newman-Keuls test or with Student's t-test, using 95% confidence (α = 0.05).
Validation of Platelet Counts on Hematology Analyzer
In this study, platelets in three PRP preparations were counted on the Cell-Dyn 3700 automated hematology analyzer and manually under a microscope at 400× magnification (n = 21 for each preparation). There was an average difference of 1.1% between the manual count and the hematology analyzer counts (Table 1). The statistical evaluation demonstrated that there was no significant difference found between the manual and automated platelet counts (P > 0.05). Furthermore, in this study, the average coefficient of variation for the Cell-Dyn counts (2.9%) was lower than the variation for the manual counts (7.1%), suggesting that the Cell-Dyn 3700 can be used to count platelets in PRP samples with greater reproducibility than manual counting.
Increase in Linear Range for Hematology Analyzer
The manufacturer limit for the linear range for platelet counts of the Cell-Dyn 3700 is 2,000 × 103 platelets/μL. To demonstrate linearity in ranges higher than the manufacturer limit, concentrated PRP samples were counted at full strength, then diluted 1:2 and 1:4 and counted again. In Figure 1, the averaged full strength platelet count is compared with the 1:2 dilution multiplied by 2 and the 1:4 dilution multiplied by 4. Statistical analysis indicated no statistical difference between the full strength and diluted samples multiplied by the corresponding dilution factor (P > 0.05). In addition, the linear correlation value between the full strength and the diluted platelet counts was found to be R2 = 0.96 (data not shown). These results suggest that no dilution is necessary to count PRP samples with concentrations up to approximately 4,800 × 103 platelets/μL.
Validation of Platelet Counts with Citrate Anticoagulated Blood
The Cell-Dyn 3700 is manufacturer-validated only with blood anticoagulated with EDTA. Human blood was drawn into EDTA-coated Vacutainer tubes (no dilution of whole blood) or was drawn into 7 mL uncoated Vacutainer tubes with 0.8 mL of ACD-A added (dilution factor 1.11). Because PRPs could not be prepared using the same EDTA, only baseline blood samples were compared. Figure 2 summarizes the results for the platelet counts between EDTA and ACD-A anticoagulated human blood. The average platelet count for the EDTA blood for all three individuals was 220,000 platelets/μL ± 17,000 platelets/μL, whereas the average platelet count for the equivalent ACD-A blood multiplied by the dilution ratio was 214,000 platelets/μL ± 17,000 platelets/μL. These values were found to be statistically the same (P > 0.05). Therefore, the Cell-Dyn is capable of counting ACD-A anticoagulant blood accurately. If results are to be compared with other forms of anticoagulated blood, the dilution factor must be considered.
Resuspension and Sample Preparation of PRP for Accurate Counts
Adequate resuspension of the PRP and correct sample preparation is required for accurate platelet counts. A time course of platelet counts while on a rocker to aid in resuspension is shown in Figure 3. In all three samples, the average platelet count was significantly less when the sample was measured immediately after preparation of the PRP than when the sample was resuspended on the rocker (P < 0.05). After 5 minutes on the rocker, the platelet counts increased and remained constant throughout the 20 minute time course (P > 0.05). These results demonstrate the importance of agitation of the PRP sample for at least 5 minutes on a rocker before attempting to perform a platelet count.
When a 0.6 mL sample was extracted from the PRP after shaking the disposable but before resuspension on the rocker and counted, the platelet count was significantly less than the same full sample resuspended (Fig 4). The average platelet count for the 0.6 mL samples was 1,958 × 103 platelets/μL ± 140 × 103 platelets/μL as opposed to 2,093 × 103 platelets/μL ± 56 × 103 platelets/μL for the full sample (P = 0.007). These data support the premise that an accurate platelet count cannot be guaranteed when a portion of the PRP is removed for counting. However, in another experiment, it was demonstrated that the samples could be divided in half and accurately counted after the entire PRP sample was allowed to resuspend on the rocker for 15 minutes (data not shown).
Platelet Counts for PRP Created by GPS II
Both bovine and human PRPs were created with the GPS II, and fold increases in platelet concentration and percent recovery of whole blood platelets were calculated. With the bovine blood, the average baseline whole blood count was 328,000 platelets/μL ± 69,000 platelets/μL. After concentration with the GPS II, the samples were concentrated to an average 2,645,000 platelets/μL ± 680,000 platelets/μL (Fig 5). The average fold increase for the 153 bovine PRP samples created with the GPS II was 8.06 ± 1.14, and the average percent recovery for the same samples was 75.7% ± 9.3%. Table 2 gives the results of the human platelet counts. The average baseline platelet count for all three human samples was 194,000 platelets/μL ± 59,000 platelets/μL and the PRP platelet count for the three samples was 1,845,000 platelets/μL ± 353,000 platelets/μL. The average fold increase as a result was 9.7 ± 1.1 and the average percent recovery was 91.2% ± 5.5%. These data illustrate that the GPS II can concentrate both human and bovine blood effectively, and both can accurately be counted using the current sample preparation steps described in this study.
To determine the efficacy of PRP, it is first necessary to accurately quantitate the platelet concentration. Only then can the number of platelets required for each surgical application be addressed. Hematology analyzers are routinely used to accurately and precisely count elements within blood. However, most systems are designed to operate in ranges found within whole blood. Therefore, counting platelets in concentrations found in PRP can generate errors or spurious results. There are mechanical limitations in both the hematology analyzers and the specimen preparation steps that can affect the accuracy of platelet counts. The purpose of this study was to validate one type of hematology analyzer for counting platelets in PRP and to define the specimen preparation methods needed to ensure accuracy. The validations performed here were designed for this particular counter, and validation for other hematology analyzers would need to be designed according to the device's specific operational parameters and limits.
Both bovine blood and human blood were used to generate PRPs. The platelets were counted on the Cell-Dyn 3700. Through manual counting techniques, it was apparent that the GPS II adequately produced bovine PRP and that the fold increase from baseline blood was similar to that seen in human PRP preparations. In addition, the Cell-Dyn 3700 is equipped with a veterinary package, allowing the system to be used for multiple species. For hematology analyzers to count platelets of any species, the machine must be able to discern RBCs from platelets according to size. Human RBCs have a mean cell volume of approximately 79 to 97 fL, whereas the platelets have a mean platelet volume of 5.6 to 10.4 fL.15 Bovine RBCs and platelets are similarly sized, with a mean cell volume of 40 to 60 fL and a mean platelet volume of approximately 7 fL.16 The size difference between the RBCs and platelets is large enough for the Cell-Dyn to discern between the two cell types of particles. This is not necessarily the case for other species. For instance, goat RBCs are only 23 fL, whereas their platelets are 4.4 to 8.3 fL.15 In a preliminary study, we were unable to count goat platelets from a PRP using the Cell-Dyn 3700. Our counts were consistently falsely high, presumably because the system was counting both RBCs and platelets as platelets (data not shown). It is assumed that for the purposes of this study, PRP preparations with either human or bovine blood were equivalent in platelet concentration and within capacity of the Cell-Dyn 3700 to generate accurate counts.
The first validation step was to evaluate the Cell-Dyn 3700 platelet counts in PRPs as compared with manual counts. This technique was used by Veillon et al17 to evaluate the accuracy of several different hematology analyzer systems by different manufacturers. In this study, the platelet counts of the PRPs were statistically equivalent to the manual counts. The Cell-Dyn was also capable of more precision, as evident by lower coefficient of variation than the manual counts. In addition to being able to count platelets in PRPs accurately, the Cell-Dyn 3700 is equipped to provide a warning flag if platelet clumps are detected. Counts with platelet flag clumping errors were discarded, and new PRP samples were counted.
The Cell-Dyn 3700 was validated by the manufacturer to count platelets up to 2,000 × 103 platelets/μL, a level the GPS PRP preparation can exceed. It is important to understand the linear limit of the hematology analyzer to determine whether specimen dilution is necessary for that particular system. Concentrated PRPs were created and counted at full strength and at a 1:2 (50%) and 1:4 (25%) dilution. When the counts were plotted and a linear regression analysis was performed, a R2 = 0.96 correlation was found. Furthermore, as shown in Figure 1, when the diluted platelet counts were multiplied by the dilution factor, the values were found to be statistically equivalent to the undiluted samples. Taken together, these results suggest that the Cell-Dyn 3700 can be used to count PRP up to approximately 4,800 × 103 platelets/μL without a dilution step.
The last system parameter validated was platelet counts with blood anticoagulated with ACD-A. Citrate-based anticoagulants in liquid form are often used in blood bags and in stored platelet populations because they can maintain platelet viability and metabolism better than EDTA.14,18 Most platelet concentration devices recommend the use of citrate anticoagulated blood with their systems. However, the Cell-Dyn's manufacturer only validated the system with blood anticoagulated with EDTA. This anticoagulant is coated onto the inside of Vacutainer tubes and does not significantly dilute the blood. Our results demonstrated that the platelet counts in ACD-A anticoagulated blood, once multiplied by the dilution factor, were equivalent to counts for samples from the same donor anticoagulated with EDTA. Again, each anticoagulant used should be validated on the system being used for platelet counts. PRPs could not be compared between the two anticoagulants because EDTA tubes are not available in a large enough volume.
After determining that the counting system would function appropriately, focus was then placed on the requirements for sample preparation of the platelet concentrate. As demonstrated by the time course of the PRP on the rocker, at least 5 minutes of agitation ensures that the PRP is adequately mixed. When the platelets are first removed from the centrifuge and resuspended by 30 seconds of shaking, the platelets are not evenly distributed and can clump. This clumping inhibits result in full platelet activation, as demonstrated by low p-selectin expression on the PRP platelets.1 Without providing time for the platelet clumps to relax, low platelet counts and platelet error flags on hematology analyzers can occur.
A common practice used in the surgical arena is to collect a PRP for a patient and then to aspirate off a small fraction for counts. The remaining PRP is given back to the patient for therapeutic use. As demonstrated in this study, sample fractionating cannot provide accurate platelet counts. After PRP preparation, the platelets are not evenly distributed and may be clumped. In this study, we demonstrated that a 0.6 mL fraction gave a statistically significant lower platelet count than the same full sample allowed to rest on the shaker for 15 minutes. The smaller fraction of PRP before shaking on the rocker could have lower platelet counts either because platelets in the disposable may be slightly clumped or the platelets may not be evenly distributed throughout the entire PRP volume. Therefore, this study suggests that for accurate platelet counts to be achieved, the entire PRP sample must be removed and allowed to sit on the rocker at least 5 minutes before counting.
In addition, this study demonstrated that PRP samples should be mixed immediately before being counted. This study suggests that using manual modes in the hematology analyzers are more beneficial than automatic counting modes to prevent tubes of PRP having time to settle before counting. The automatic counting mode on the hematology analyzer can allow the blood sample to settle for some time before counting occurs. Manual counting methods ensure that the blood sample is agitated up to the moment it is counted. The amount of platelet settling and resuspension may vary depending on the RBC percentage in the PRP. PRP preparation systems can vary in the amount of platelets concentrated and the amount of RBCs that are removed, so each platelet concentrate system could require an individualized sample preparation and resuspension protocol.
Once the hematology analyzer and the PRP preparation device are validated, then accurate platelet counts can be achieved. As shown in Figure 5 and Table 2, the GPS II can provide at least an eightfold increase in platelet concentration over baseline, with a greater than 70% platelet recovery. These data are consistent with a 10 patient clinical study that analyzed platelet counts and growth factor content in platelets collected with the GPS System.1
The many issues that can preclude accurate platelet counts from a PRP using an automated machine have been demonstrated in this study. With several platelet concentration devices commercially available, comparison of the outputs of these devices is inevitable.19,20 However, without accurately validating the hematology analyzer and preparing the PRP samples appropriately for each system being compared, the results may be inaccurate and misleading.
PRPs have demonstrated numerous clinical benefits to patients.21-25 There are many devices on the market that concentrate platelets to differing levels. At this time, the amount of platelets that is most efficient for each surgical application is unclear. Marx11 recommended that at least 1,000 × 103 platelets/μL are generally required. The GPS system can provide platelets up to eight times baseline concentration,1 and the normal human platelet range is 200 × 103 platelets/μL to 400 × 103 platelets/μL. This implies that the GPS provides platelets in a range of 1,600 × 103 platelets/μL to 3,200 × 103 platelets/μL. To further understand the amount of platelets required for each application, adequate platelet counting protocols must be established for platelets in the ranges discussed above. This study addresses the issues in ensuring the hematology analyzer used can accurately count the platelets in the PRPs and that the PRPs are prepared in such a way as to provide accurate counts. Only after these steps have been followed can comparison between devices be made.
1. Eppley BL, Woodell JE, Higgins J. Platelet quantification and growth factor analysis from platelet-rich plasma: Implications for wound healing. Plast Reconstr Surg 2004;114:1502-1508
2. Weibrich G, Kleis WK, Buch R, Hitzler WE, Hafner G. The Harvest Smart PRePTM system versus the Friadent-Schutze platelet-rich plasma kit. Clin Oral Implants Res 2003;14:233-239
3. Kevy SV, Jacobson MS, Blasetti L, Fagnant A. Preparation of growth factor enriched autologous platelet gel. Soc Biomat 2001;24:262
4. Slater M, Patava J, Kingham K, Mason RS. Involvement of platelets in stimulating osteogenic activity. J Orthop Res 1995;13:655-663
5. Spencer ME, Tokunaga A, Hunt TK. Insulin-like growth factor binding protein-3 is present in the alpha-granules of platelets. Endocrinology 1993;132:996-1001
6. Mishra A, Pavelko T. Treatment of severe lateral epicondylitis with platelet rich plasma. Am Acad Orthoped Surg 2005
7. Lowery GL, Kulkarni S, Pennisi AE. Use of autologous growth factors in lumbar spinal fusion. Bone Suppl 1999;25:47S-50S
8. Aminiam B, Shams M, Soveyd M, Omrani GhR. Topical autologous platelet-derived growth factors in the treatment of chronic diabetic ulcers. Arch Iranian Med 2000;8
9. Bhanot S, Alex JC. Current applications of platelet gels in facial plastic surgery. Facial Plast Surg 2002;18:27-33
10. Garg AK. The use of platelet-rich plasma to enhance the success of bone grafts around dental implants. Dent Implantol Update 2000;11:17-21
11. Marx RE. Platelet-rich plasma: Evidence to support its use. J Oral Maxillofac Surg 2004;62:489-496
12. Wertz RK, Koepke JA. A critical analysis of platelet counting methods. Am J Clin Pathol 1977;68(1 Suppl):195-201
13. Marx RE. Platelet concentrate: A strategy for accelerating and improving bone regeneration. In: Davies JE, ed. Bone Engineering. Toronto: University of Toronto; 2000:447-453
14. Marx RE. Platelet-rich plasma (PRP): What is PRP and what is not PRP? Implant Dent 2001;10:225-228
15. Lewis JH. Comparative Hemostasis in Vertebrates. New York: Plenum Press, 1996;316
16. Jain NC. Essentials of Veterinary Hematology. Philadelphia: Lippincott, Williams, and Wilkins, 1993;415
17. Veillon DM, Curry S, Tubbs K, Kaltenbach J, Cotelingam JD. Decision dilemma: How we chose a new hematology analyzer. Lab Hematol 2000;6:151-156
18. Landesberg R, Roy M, Glickman RS. Quantification of growth factor levels using a simplified method of platelet-rich plasma gel preparation. J Oral Maxillofac Surg 2000;58:297-300
19. Kevy SV, Jacobson MS. Comparison of methods for point of care preparation of autologous platelet gel. JECT 2004;36:28-35
20. Waters JH, Roberts KC. Database review of possible factors influencing point-of-care platelet gel manufacture. J Extra Corpor Technol 2004;36:250-254
21. Margolis DJ, Kantor J, Santanna J, Strom BL, Berlin JA. Effectiveness of platelet releasate for the treatment of diabetic neuropathic foot ulcers. Diabetes Care 2001;24:483-488
22. Marx RE, Carlson ER, Eichstaedt RM, Schimmele SR, Strauss JE, Georgeff KR. Platelet-rich plasma: Growth factor enhancement for bone grafts. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 1998;85:638-646
23. Mooar PA, Gardner MJ, Klepchick PR, Sherk HH. The efficacy of autologous platelet gel in total knee arthroplasty: An analysis of range of motion, hemoglobin, and narcotic requirements. Am Acad Orthoped Surg 2000;PE148
24. Welsh W. Autologous platelet gel-clinical function and usage in plastic surgery. Cosmetic Dermatology 2000;13-18
25. Whitman DH, Berry RL, Green DM. Platelet gel: an autologous alternative to fibrin glue with applications in oral and maxillofacial surgery. J Oral Maxillofac Surg 1997;1294-1299
© 2005 Mutaz B. Habal, MD