Secondary Logo

Journal Logo

Subcutaneous Injection of Tranexamic Acid to Reduce Bleeding During Dermatologic Surgery

A Double-Blind, Placebo-Controlled, Randomized Clinical Trial

Zilinsky, Isaac, MD*; Barazani, Tami Brutman, PhD; Visentin, Denis, PhD; Ahuja, Kiran, PhD; Martinowitz, Uriel, MD; Haik, Josef, MD§,║

doi: 10.1097/DSS.0000000000001786
Original Article
Free

BACKGROUND Topical application, oral, and IV injection of tranexamic acid (TXA) have been used to reduce surgical bleeding.

OBJECTIVE To evaluate the safety and efficacy of TXA injected subcutaneously to reduce bleeding during dermatologic surgery.

METHODS In this double-blinded, placebo-controlled, randomized prospective study, 131 patients were randomized to subcutaneous injection of lidocaine 2% diluted 1:1 with either saline (placebo) or TXA 100 mg/1 mL before surgery. Before the second stage or closure, size measurements of bloodstain impregnation on Telfa and surgical wound size were recorded and analyzed using mixed-effects linear regression. Subjective evaluation of hemostasis was performed using 4-point scale grading and analyzed using Fischer's exact test.

RESULTS One hundred twenty-seven patients completed the study. The bloodstain to surgical wound size ratio was smaller in the TXA group (1.77) compared with the placebo group (2.49) (p < .001). An improved effect of TXA on bleeding was observed in the subgroup of patients receiving anticoagulants (mean difference; 95% confidence interval; −0.83; −1.20 to −0.46 p < .001). The subjective hemostasis assessment was significantly better in the TXA group overall (p = .043) and anticoagulant subgroup (p = .001) compared with the placebo group.

CONCLUSION Subcutaneous injection of TXA was safe, reduced bleeding during dermatologic surgery, and particularly effective for patients receiving anticoagulation treatment.

*Mohs Unit, Sheba Medical Center, Tel Hashomer, Affiliated to Sackler Faculty of Medicine, Tel Aviv University, Tel Aviv, Israel;

The National Hemophilia Center and Thrombosis Institute, Sheba Medical Center, Tel Hashomer, Affiliated to Sackler Faculty of Medicine, Tel Aviv University, Tel Aviv, Israel;

School of Health Sciences, University of Tasmania, Launceston, Tasmania, Australia;

§Burns Unit, The Department of Plastic and Reconstructive Surgery, Sheba Medical Center, Tel Hashomer, Affiliated to Sackler School of Medicine, Tel Aviv University, Israel;

Burn Injury Research Node, Institute for Health Research, University of Notre Dame Fremantle, Fremantle, Western Australia, Australia

Address correspondence and reprint requests to: Josef Haik, MD, Burns Unit, The Department of Plastic and Reconstructive Surgery, Sheba Medical Center, Tel Hashomer, affiliated to Sackler School of Medicine, Tel Aviv University, Tel Aviv, 52621, Israel, or e-mail: josek.haik@gmail.com

The authors have indicated no significant interest with commercial supporters.

Bleeding during dermatologic surgery is an inevitable part of the procedure. Its intensity is dictated by various factors such as surgery location, size and depth of the surgical wound, involvement of larger blood vessels, presence of high blood pressure, and the use of medications that inhibit clotting. Dermatologic surgical procedures such as Mohs micrographic surgery (MMS) are usually safe and are not characterized by excessive blood loss; nevertheless, even a small amount of bleeding may obscure the surgical field and result in hematomas in the postoperative period. Attempts to minimize surgical bleeding include better control of blood pressure, local injection of adrenaline with anesthetic agent, using an electrocoagulation device during surgery, and discontinuing anticoagulants before surgery. Excessive electrocoagulation contributes to tissue damage and may result in delayed wound healing, a higher tendency for infection, and a less than optimal scar.1,2 Discontinuing the use of anticoagulants before surgery is dangerous and may result in thromboembolic events such as stroke.3,4

Antifibrinolytic drugs promote blood clotting by preventing the blood clot from breaking down. One of the commonly used antifibrinolytic drugs is tranexamic acid (TXA) (trans-4-aminomethyl cyclohexane carboxylic acid) (TXA, Cyklokapron), a synthetic lysine analogue that reversibly blocks the binding sites of plasminogen and prevents plasminogen activation to plasmin and the lysis of polymerized fibrin in the blood clot.5 Tranexamic acid is low cost, readily available in a surgical setting, and has been used routinely for many years to reduce bleeding both during and after surgery, and can be administered using IV, orally or topically. If there are no contraindications, TXA can be safely administered postoperatively to reduce healing time and minimize complications of bleeding such as bruising. It is used topically during tooth extraction, orthopedic procedures, cardiac surgery, and other surgical procedures.6–9 Tranexamic acid is also used intravenously to reduce the need for blood transfusion in orthopedic,10 cardiac,11 and liver transplantation,12 for children undergoing tonsillectomy13 and to reduce the risk of rebleeding within 24 hours after a subarachnoid hemorrhage.14 Tranexamic acid is also used to reduce blood loss after vaginal delivery and caesarian section when given as prophylaxis.15

Tranexamic acid dosage varies widely, according to the literature and in clinical practice. Tranexamic acid has a half-life elimination of 2 to 11 hours with 95% of the drug eliminated unchanged through the urine. The safety of TXA has been demonstrated in numerous publications. A systematic review that included 20,451 patients concluded that TXA reduces mortality in trauma patients with bleeding, without increasing the risk of adverse reactions such as vascular occlusive events (myocardial infarction, stroke, deep vein thrombosis, and pulmonary embolism).16,17 A single report suggests that high-dose TXA (≥100 mg/kg) is independently associated with an increased risk of early seizure in open heart surgery.18 There is one case report of accidental intrathecal administration of a large dose of TXA resulting in convulsions and refractory ventricular fibrillation19 and a case report of polymyoclonus seizure resulting from accidental intrathecal injection of TXA in spinal anesthesia.20

In this study, we investigated the safety of the hemostatic–anesthetic mixture (containing TXA) injected subcutaneously and the efficacy of TXA in reducing bleeding. To assess the amount of bleeding from the wound, we used a normalized measure of bloodstain size to surgical wound size ratio (Figure 1) as has been reported in our previous article.21 The effect of TXA on clot stabilization and reduction of bleeding in surgery and trauma is well documented in the medical literature but is only described in oral consumption, topical application, and IV injection. To the best of our knowledge, subcutaneous or intradermal injection of TXA for the purpose of intraoperative and postoperative bleeding reduction has not previously been reported.

Figure 1

Figure 1

Back to Top | Article Outline

Materials and Methods

This is a single-center, double-blind, prospective, randomized, clinical study. The protocol was approved by the local ethics committee at the Sheba Medical Center (approval number SMC-1079-14). The study was conducted in accordance with the declaration of Helsinki, and written informed consent was obtained from all participants before their undergoing surgery.

All patients who were referred to MMS for head and neck skin cancer were offered to participate in this study. Exclusion criteria were systolic blood pressure greater than 200 mmHg, surgical wound size greater than 4 cm, unwillingness of patient to sign the informed consent, a known allergy to TXA, and a known history of personal or familial coagulation disorder. Patients receiving anticoagulants were not excluded, and note of the drug type and whether or not it was discontinued before surgery was recorded. Patient stopping acetylsalicylic acid for less than 5 days was considered as still taking acetylsalicylic acid. Before surgery the following data were collected: demographic data, concurrent medical conditions, smoking status, and systemic medications (specifically, anticoagulation treatment and when the last dose had been taken).

One hundred fifty patients were assessed for eligibility with the aim to enroll a minimum of 100 participants. Nineteen patients did not meet the inclusion/exclusion criteria. One hundred thirty-one patients who participated in the study were randomized into either the TXA group or the placebo group (Table 1). All the surgical procedures and measurements were performed by a single surgeon. All the dressings were performed by the same nurse. The patient blood pressure was recorded before each MMS stage, and the pulse was monitored continuously with a pulse oximeter. Mohs micrographic surgery was performed under local anesthesia. After marking the skin, the surgeon injected each patient subcutaneously with a mixture of 2% lidocaine with either TXA 100 mg/1 mL (TXA group) or normal saline 0.9% sodium chloride (placebo group) in a 1:1 mixture. This created a final mixture with an effective 1% lidocaine as an anesthetic agent in both groups and a TXA concentration of 50 mg/mL in the TXA group. The TXA solution was mixed with lidocaine in the same syringe. We have presented in our previous study that the mixture (TXA + lidocaine) does not interfere with the antifibrinolytic action of TXA.21 Adrenaline was not added to the anesthetic solution during the study phase.

TABLE 1

TABLE 1

The different mixtures were prepared in coded identical 1-mL insulin syringes by the study nurse before surgery. There were an equal number of syringes in the 2 groups. The surgeon randomly chose the syringe for each case. The coding list was opened only after the completion of the study. The surgeon and patient were unaware of the type of mixture that was used. The mixture was injected subcutaneously until the patient reported no pain when checking with a needle prick. The total amount injected varied according to the tumor size and patient individual pain tolerance. Average volume injected was between 1 and 6 mL. To allow for a significant effect of the mixture, the surgeon waited 15 minutes before starting the surgery. After the first MMS stage, a routine hemostasis was performed with a standard electrocautery device (Bovie) with a Colorado tip, set up to level 20 kHz strength, and used directly on points of bleeding or through fine jeweler's forceps where small arterioles were nicked. The electrocoagulation was performed using short pulses until bleeding was fully controlled. The surgical wound was covered with a 6 × 6 Telfa AMD pad attached to the wound and 1 folded 10 × 10 gauze on top of that, secured with a Micropore tape. The patient was then transferred to the waiting area until the second MMS stage or reconstruction procedure. After 30 minutes, the patient was returned to the surgical unit, and the dressing was removed. At this stage, 2 measurements were taken and recorded; the long axis of the surgical wound and the long axis of the bloodstain impregnation on the Telfa pad (Figure 1). This model for measuring a small amount of bleeding is more accurate than gauze weighing because of the difference in order of magnitude between the weight of the blood soaked in the gauze and the gauze weight itself.

Subjective evaluation of hemostasis was performed by a single surgeon using a 4-point ordinal scale as: excellent: better than expected/predicted in this type of procedure; good: as expected in this type of procedure with minimal oozing; moderate: less than optimal for the type of procedure with moderate oozing but no need to add gauze dressing; and poor: pinpointed oozing without a presence of a pumping vessel that needed adding gauze dressing and some local pressure.22 The same surgeon also recorded whether he believed the solution contained TXA or not, based on his subjective assessment of whether he saw a blood clot in the surgical wound and on the 4-point scale. From this stage on, the surgery resumed routinely as is customary in our practice, at the second stage of the Mohs surgery or repair: bupivacaine (5 mg/mL) with adrenaline (final concentration of 1:200,000) was injected to both study arms (1–3 mL).

Analyses were performed on the transformed (inverse) data because the raw data for bloodstain to surgical wound size ratio were not normally distributed. The comparative results are shown as mean difference with 95% confidence intervals (CIs) but p-values are from the transformed variable. Graphs were plotted using Prism 7 (GraphPad Software Inc., CA). Ratio for bloodstain size to wound size for placebo and TXA groups were compared using mixed-effect linear regression (Stata, v. 14; StataCorp, TX). All data were analyzed with and without adjusting for any effect of confounders including age and sex. Data for surgeon's assessments of hemostasis and type of intervention were analyzed using Fischer's exact test. Subgroup analyses were performed between the 2 interventions to compare participants taking anticoagulants and those not taking anticoagulants.

Back to Top | Article Outline

Results

Data were collected between November 10, 2014, and December 19, 2016. One hundred thirty-one patients took part in the study (67 in the placebo and 64 in the TXA group). Four patients were excluded from data analysis (Figure 2). In 3 cases, a small pumping arteriole was detected at the time of dressing removal that obscured proper measurement of perfused bleeding. In one patient, an allergic reaction was suspected because of redness and local edema that was noticed 5 minutes after the injection. The patient was excluded from the study and reported by the study nurse to be injected with TXA. The local signs disappeared spontaneously after 2 hours without further sequelae.

Figure 2

Figure 2

Baseline characteristics of the study participants are presented in Table 1. The surgical wound size for the TXA group was significantly larger than the placebo group. There was no difference in the bloodstain size between the 2 groups. The ratio of bloodstain to surgical wound size was smaller in the TXA group compared with the placebo group (mean difference (MD); 95% CI; −0.72; −1.09 to −0.35; p < .001) (Table 2). These comparative results did not change after controlling for the potential confounders of age and sex.

TABLE 2

TABLE 2

A greater effect of TXA on the bloodstain to surgical wound size ratio was observed in the anticoagulant subgroup. Patients on anticoagulants in the TXA group had significantly lower bloodstain size to surgical wound size ratio than those in the placebo group (MD; 95% CI; −0.83; −1.20 to −0.46 p < .001). Similarly, patients not on anticoagulants (anticoagulant free) in the TXA group had significantly lower bloodstain size to surgical wound size ratio than those in the placebo group (MD; 95% CI; −0.60; −1.18 to −0.01 p = .045) (Figure 3).

Figure 3

Figure 3

The hemostasis assessment performed by the surgeon (including all patients) using the 4-point scale was significantly better (p = .043), with larger number of excellent and good assessments in the TXA group compared with the placebo group (Figure 4). Furthermore, in the anticoagulant subgroup, the hemostasis evaluation was significantly better (p = .001), with larger number of excellent and good assessments for the TXA group (92.8%) compared with the placebo group (57.2%). However, no significant differences were observed for the anticoagulant-free subgroup (Table 3). The surgeon's assessment of whether the patient was on TXA or placebo was correct more often (p = .001) for the TXA group (63.3%) than for the placebo group (32.8%). Furthermore, the subgroup analysis showed the surgeon's assessment of TXA/placebo was correct more often for the TXA group than for the placebo group in both the anticoagulant (p = .008) and nonanticoagulant (p = .038) subgroups.

Figure 4

Figure 4

TABLE 3

TABLE 3

Back to Top | Article Outline

Discussion

In this study, TXA solution injected subcutaneously together with lidocaine before MMS, demonstrated a significant hemostatic effect compared to lidocaine with saline. The decrease in bloodstain to defect ratio and significantly improved hemostasis assessment for subcutaneous TXA injection is due reversibly blocking plasminogen-binding sites preventing activation to plasmin and lysis of polymerized fibrin in clotting. These results were more pronounced in patients taking anticoagulation drugs such as acetylsalicylic acid (aspirin), warfarin (Coumadin), apixaban (Eliquis), clopidogrel (Plavix), enoxaparin (Clexane), and rivaroxaban (Xarelto). The explanation for this phenomenon may be shifting of the equation, bleeding to clot formation, to the right regardless the anticoagulant that was taken (Figure 5). Although blood clotting is a normal phenomenon in patients who are not taking anticoagulants, the effect of antifibrinolytic agents are more pronounced in patients on anticoagulants with impaired clotting cascades.

Figure 5

Figure 5

Subcutaneous injection of lidocaine with TXA was well tolerated in our study with only one reported adverse outcome, a mild local allergic reaction that quickly resolved. No thromboembolic events were noted. Previous studies have suggested not to discontinue anticoagulation treatment during dermatologic surgery despite an increase in bleeding to minimize the possible risk of serious thromboembolic complications.23 This work presents a method to minimize the increased bleeding in patients under anticoagulation treatment by locally propagating coagulation effect.

Dermatologic surgeons are trying constantly to reduce pain during local anesthesia, using different injection techniques24 and routinely buffering the anesthetic solutions with bicarbonate.25,26 Hence, an additional benefit of TXA may be its buffering effect. With a pH of between 6.5 and 8,27,28 TXA may serve as an alternative to using bicarbonate for buffering acidic local anesthetic solutions, with the added effect of hemostasis. Adding TXA to the tumescent mixture may also be beneficial in local anesthesia for liposuction.

Although adrenaline was not used during the experimental phase of this study, TXA and adrenaline effect on bleeding control is complementary. Adrenaline reduces bleeding by immediate constriction of blood vessels, and TXA stabilizes the clot, thus enhancing hemostasis. Adrenaline is relatively short acting, and some patients tend to bleed after its effect dissipates. The effect of TXA, however, continues and prevents later onset of bleeding and subcutaneous hematomas. It is possible that a mixture of lidocaine or bupivacaine with adrenaline and TXA will be more effective in controlling bleeding, and this combination efficacy and safety should be explored in future studies. The literature has identified TXA as a skin lightening agent when used repeatedly;29–31 however, in our study, no cases of hypopigmentation were observed in a follow-up examination after 3 months.

Because TXA has been shown to inhibit angiogenesis and neovascularization in tumors in an animal model,32 there was a theoretical risk that TXA could interfere with skin grafts take or flaps survival. We did not observe any difference in outcomes of reconstruction with skin grafts or flaps at the suture removal after 1 week; however, this should be monitored carefully in similar future studies.

The results of this study may be biased by some factors: location and depth of the surgical wound, the intensity of the cautery used, momentary high peaks of blood pressure while the patient waits between the Mohs stages (coughing, sneezing, and valsalva in the toilet), hormonal factors in women (menstruation and pregnancy). Nevertheless, we have no reason to expect a different distribution of these factors between the relatively large randomized study groups. Future studies should consider potential issues with blinding in the research design. The effect of buffering of the TXA solution on blinding was not considered in this study design, which may provide observable variations in pain during injection between the 2 groups. It would also be advisable to have more than one surgeon perform the assessments, which would also minimize issues with blinding and provide interoperator reliability measures. The efficacy of subcutaneous injection of the combined solution described herein can be replicated and evaluated easily by any dermatologic surgeon.

Back to Top | Article Outline

Conclusions

The results of the current study indicate that subcutaneous injection of TXA may reduce bleeding during and after dermal surgery. The hemostatic–anesthetic mixture proved to be safe and even more effective in patients receiving anticoagulation treatment. The current study supports the increasing tendency among dermatologic surgeons to avoid discontinuation of anticoagulation drugs before surgery. Instead, we present a method to minimize the increased bleeding by local enhancement of the coagulation effect. The use of TXA together with the anesthetic agent allows for better control of bleeding during and after surgery and potentially less burning sensation during injection. We suggest the use of TXA together with the anesthetic drug routinely in dermatologic surgery.

Back to Top | Article Outline

Acknowledgments

The authors thank Dr. Abigail Waldman from Harvard Medical School for manuscript review and Dr. Oded Sagiv from Goldschleger Eye Institute at Sheba Medical Center for conducting a parallel research study in ophthalmic surgery.

Back to Top | Article Outline

References

1. Lawrenson K, Stephens FO. The use of electrocutting and electrocoagulation in surgery. Aust N Z J Surg 1970;39:417–21.
2. Rappaport WD, Hunter GC, Allen R, Lick S, et al. Effect of electrocautery on wound healing in midline laparotomy incisions. Am J Surg 1990;160:618–20.
3. Sibon I, Orgogozo JM. Antiplatelet drug discontinuation is a risk factor for ischemic stroke. Neurology 2004;62:1187–9.
4. Wagner ML, Khoury JC, Alwell K, Rademacher E, et al. Withdrawal of antithrombotic agents and the risk of stroke. J Stroke Cerebrovasc Dis 2016;25:902–6.
5. Ortmann E, Besser MW, Klein AA. Antifibrinolytic agents in current anaesthetic practice. Br J Anaesth 2013;111:549–63.
6. Sa-Ngasoongsong P, Wongsak S, Chanplakorn P, Woratanarat P, et al. Efficacy of low-dose intra-articular tranexamic acid in total knee replacement; a prospective triple-blinded randomized controlled trial. BMC Musculoskelet Disord 2013;14:340.
7. Ker K, Beecher D, Roberts I. Topical application of tranexamic acid for the reduction of bleeding. Cochrane Database Syst Rev 2013;7:CD010562.
8. Chang CH, Chang Y, Chen DW, Ueng SW, et al. Topical tranexamic acid reduces blood loss and transfusion rates associated with primary total hip arthroplasty. Clin Orthop Relat Res 2014;472:1552–7.
9. Albirmawy OA, Saafan ME, Shehata EM, Basuni AS, et al. Topical application of tranexamic acid after adenoidectomy: a double-blind, prospective, randomized, controlled study. Int J Pediatr Otorhinolaryngol 2013;77:1139–42.
10. McNicol ED, Tzortzopoulou A, Schumann R, Carr DB, et al. Antifibrinolytic agents for reducing blood loss in scoliosis surgery in children. Cochrane Database Syst Rev 2016;9:CD006883.
11. Henry DA, Carless PA, Moxey AJ, O'Connell D, et al. Anti-fibrinolytic use for minimising perioperative allogeneic blood transfusion. Cochrane Database Syst Rev 2011;3:CD001886.
12. Molenaar IQ, Warnaar N, Groen H, Tenvergert EM, et al. Efficacy and safety of antifibrinolytic drugs in liver transplantation: a systematic review and meta-analysis. Am J Transpl 2007;7:185–94.
13. Chan CC, Chan YY, Tanweer F. Systematic review and meta-analysis of the use of tranexamic acid in tonsillectomy. Eur Arch Otorhinolaryngol 2013;270:735–48.
14. Hillman J, Fridriksson S, Nilsson O, Yu Z, et al. Immediate administration of tranexamic acid and reduced incidence of early rebleeding after aneurysmal subarachnoid hemorrhage: a prospective randomized study. J Neurosurg 2002;97:771–8.
15. Novikova N, Hofmeyr GJ. Tranexamic acid for preventing postpartum haemorrhage. Cochrane Database Syst Rev 2010;7:CD007872.
16. Ker K, Roberts I, Shakur H, Coats TJ. Antifibrinolytic drugs for acute traumatic injury. Cochrane Database Syst Rev 2015;5:CD004896.
17. Shakur H, Roberts I, Bautista R, Caballero J, et al. Effects of tranexamic acid on death, vascular occlusive events, and blood transfusion in trauma patients with significant haemorrhage (CRASH-2): a randomised, placebo-controlled trial. Lancet 2010;376:23–32.
18. Kalavrouziotis D, Voisine P, Mohammadi S, Dionne S, et al. High-dose tranexamic acid is an independent predictor of early seizure after cardiopulmonary bypass. Ann Thorac Surg 2012;93:148–54.
19. Yeh HM, Lau HP, Lin PL, Sun WZ, et al. Convulsions and refractory ventricular fibrillation after intrathecal injection of a massive dose of tranexamic acid. Anesthesiology 2003;98:270–2.
20. Mohseni K, Jafari A, Nobahar MR, Arami A. Polymyoclonus seizure resulting from accidental injection of tranexamic acid in spinal anesthesia. Anesth Analg 2009;108:1984–6.
21. Zilinsky I, Barazani TB, Shenkman B, Weisman O, et al. Topical hemostatic-anesthetic solution to reduce bleeding during Mohs micrographic surgery: a case control study. J Drugs Dermatol 2016;15:851–5.
22. Gill JC, Shapiro A, Valentino LA, Bernstein J, et al. von Willebrand factor/factor VIII concentrate (Humate-P) for management of elective surgery in adults and children with von Willebrand disease. Haemophilia 2011;17:895–905.
23. Alcalay J, Alkalay R. Controversies in perioperative management of blood thinners in dermatologic surgery: continue or discontinue? Dermatol Surg 2004;30:1091–1094; discussion 4.
24. Zilinsky I, Bar-Meir E, Zaslansky R, Mendes D, et al. Ten commandments for minimal pain during administration of local anesthetics. J Drugs Dermatol 2005;4:212–6.
25. Best CA, Best AA, Best TJ, Hamilton DA. Buffered lidocaine and bupivacaine mixture—the ideal local anesthetic solution? Plast Surg (Oakv) 2015;23:87–90.
26. MacKenzie TA, Young ER. Local anesthetic update. Anesth Prog 1993;40:29–34.
27. Frank SG, Lalonde DH. How acidic is the lidocaine we are injecting, and how much bicarbonate should we add? Can J Plast Surg 2012;20:71–3.
28. FDA. Bupivacain Hydrochloride and Epinephrine Injection, USP. USA; 2009. Contract No.: QEN-2349v1.qxp. Available from: https://www.accessdata.fda.gov. Accessed February 17, 2018.
29. Lee JH, Park JG, Lim SH, Kim JY, et al. Localized intradermal microinjection of tranexamic acid for treatment of melasma in Asian patients: a preliminary clinical trial. Dermatol Surg 2006;32:626–31.
30. Wang N, Zhang L, Miles L, Hoover-Plow J. Plasminogen regulates pro-opiomelanocortin processing. J Thromb Haemost 2004;2:785–96.
31. Nadiah H, Hannah M, Lee K, Jofrry S, et al. Use of tranexamic acid for skin whitening and melasma therapy: a product review. Arch Pharm Pract 2016;7:43–7.
32. Sundbeck A, Myrhage R, Peterson HI. Influence of tranexamic acid on tumour vascularization. An experimental microangiographic study. Anticancer Res 1981;1:295–8.
© 2019 by the American Society for Dermatologic Surgery, Inc. Published by Wolters Kluwer Health, Inc. All rights reserved.