Hair transplantation (HT) is a minimally invasive surgery in which follicular units (FU) are transferred from a donor area without alopecia to a recipient area with reduced hair density. Complications are infrequent,1 with recipient area necrosis (RN), or cutaneous tissue destruction, being one of the most feared. Its occurrence has been associated with the vascular anatomy of the scalp,2 personal history of smoking or cardiovascular risk factors, as well as aspects of the surgical technique.3,4 Its true incidence rate remains unknown to this date, and published cases are limited, with a possible influence of publication bias.5,6

In order to further investigate the triggers of necrosis and its clinical characteristics, we designed a multicenter retrospective descriptive observational study. Spanish dermatologists performing HT were invited to participate. Patients undergoing surgery using the FUE technique from 2019 through 2024, who presented at least 1 area of RN and had postoperative follow-up, were included. All patients gave their prior written informed consent. The study was conducted in full compliance with the principles outlined in the Declaration of Helsinki and with ethics committee approval (Sa-19386/19 – EC: 555). Statistical analysis was performed using SPSS (Statistical Package for the Social Sciences) v27. Quantitative variables are expressed as mean and SD, or 95%CI, and qualitative variables as valid percentages.

A total of 52 patients from 6 dermatology clinics were included: 46 men (88.5%) and 6 women (11.5%), with a mean age of 43 years (range, 25–64 years). Of these, 15.4% were smokers and 9.6% had hypertension. A total of 88.5% presented androgenetic alopecia (>90% Hamilton III). The FUE technique was used in all patients (0.8mm cylindrical punch in 95.5%). In most cases, trunk anesthesia was used, and 87.8% of patients additionally received tumescence. The maximum adrenaline concentration used was 1/100,000, and in >40% the concentration was 1/450,000 or adrenaline was not used. Incisions were performed with a sapphire blade in 43% of cases (most commonly 1.5mm diameter) and in 57% with a needle (18G most widely used). The mean incision depth was 4.5±0.5mm, with a mean density of 55±11.4FU/cm2. In 75% of patients, the temples and frontal forelock were treated. Fig. 1 illustrates the sequential clinical characteristics of necrosis. After crust removal, epidermal involvement was observed in 11.9% of cases, superficial dermis in 35.7%, deep dermis in 38%, and subcutaneous cellular tissue in 7.1%. The most frequently affected areas were the right temple (41%) and the frontal forelock (38%). The mean size of each necrotic area was 2.1cm2 (±1.76; 0.5–10.5), and the mean number of affected areas was 1.5 (±0.75; 1–3). A total of 27.6% of the patients exhibited clinical signs of exudate with negative cultures. The mean time to complete resolution was 23±6 days. Survival of follicular units in the affected area ranged between 10 and 60% in 43.8% of cases and between 60 and 90% in 47.9% of cases (Fig. 2).

Fig. 1.

Characteristics of necrosis: sequential images of the development process with percentages of occurrence of the different findings. (A) White area after anesthesia/tumescence (intraoperative). (B) Purplish area after incisions (intraoperative). (C) Area of intense reactive bleeding (intraoperative). (D) Blackish skin area (88.9% 24h after the procedure). (E) Perifollicular purulent exudate (75% 48h after the procedure). (F) Interfollicular serous exudate (58.3% 48h after the procedure). (G) Thick protruding crust (51.9% 72h after the procedure). (H) Sunken blackish eschar (50% more than 96h after the procedure). (I) Erosion after crust removal (mean onset 11±3 days). (J) Pseudotufts.

Fig. 2.

Sequelae of necrosis after resolution. (A) Whitish, atrophic, and sunken skin observed in 52.8% of cases. (B) Whitish fibrous reticulated network with and without shaving observed in 80% of cases.

RN is the result of tissue damage and compromised vascularization of the FU. Three case series have been published, totaling 40 patients, with the present study being the largest, including 52 patients (Table 1). Both the “classic” predisposing personal history3 (poorly represented in our series) and the use of adrenaline7 (maximum concentrations of 1/100,000) do not appear to be triggering factors in our patients. Although there is no consensus on the number of FU to implant per cm2, dense packing is discouraged,8 as the created recipient sites may compromise vascularization (especially if they are deep and large). In our series, the number of FU grafted per cm2 was higher vs other studies,3,4 and the frontal arc (area with highest graft density) was the most affected region vs the interparietal region reported in other series,3–5 traditionally associated with necrosis due to the anatomical vascular configuration of the scalp.9 Most of our patients exhibited >1 area of RN with a mean size of 2.1cm, whereas other series describe a single large necrotic area. Furthermore, we report higher follicular survival rates, as the depth of necrosis rarely exceeded the deep dermis. Study limitations include its retrospective nature, absence of a control group, and sample size. The high density currently demanded by patients in frontal arc HT procedures may result in superficial necrotic tissue damage in multiple small areas, giving rise to a new paradigm in the presentation of RN. It is proposed to inform patients about this potential complication, consider staged procedures, or even reduce the size of recipient sites. The detailed clinical description provided here will facilitate early diagnosis and help address patient concerns.