Factors Influencing Postoperative Hyperesthesia (Discomfort)
in Hair Restoration Surgery
Background While esthetic outcomes in hair restoration surgery (HRS) have improved markedly since the advent of follicular unit transplantation (FUT), various undesirable sequelae persist. We investigated the technical and demographic variables that may contribute to the frequency of postoperative hyperesthesia.
Methods A multicenter retrospective chart review involving 552 patients undergoing HRS from 1999 to 2009.
Results A total of 19 patients (3.4%) reported postoperative hyperesthesia in either the donor or recipient area of their scalp. Although many trends emerged, one variable significantly influenced the rate of this neurosensory symptom. While no patient who had all previous and current HRS sessions performed entirely within the same investigated surgical practice (n = 42) experienced postoperative hyperesthesia, 14% of our patients who underwent prior HRS by a physician outside of the investigated surgical group (n = 35) developed this complication (P = 0.0404). The amount of intraoperative electrocautery to maintain hemostasis (P = 0.0897), degree of tension upon donor wound closure (P = 0.3044), and extent of donor wound edge undermining (P = 0.4420) influenced the frequency of this sequela to a lesser degree.
Conclusion These results suggest that physicians planning repair sessions on patients who have undergone prior HRS by a physician at a different surgical center should include the specific caveat of increased incidences of postoperative hyperesthesia in their preoperative consultation.
Technical advances within the field of hair restoration surgery (HRS) have led to considerable improvements in the esthetic outcome and an increase in overall patient satisfaction. Despite these enhancements, various unfortunate sequelae stemming from the procedure persist. Postoperative hyperesthesia and hypoesthesia resulting from peripheral nerve injury are consequences whose avoidance has remained elusive – perhaps, in part, owing to their lack of cosmetic and systemic medical significance. The discomfort they may cause, however, can negatively impact patients’ quality of life.
In HRS, changes in scalp sensitivity typically originate from injured peripheral nerves along the donor region anywhere from nerve root to terminus. As neuronal pathway distribution within scalp subcutaneous tissue is centripetal (with larger nerve trunks located more laterally before decreasing in caliber as they pass both medially and centrally), larger nerve bundles are in jeopardy of being transected during classical donor strip removal, the harvesting technique used by 88.5% of practicing surgeons [Fig. 1].1–5 Traumatic axonal damage may result from laceration, compression, traction, direct injection, or an electric current.6 Ischemia and infection may also indirectly lead to this consequence.6 In a limited number of patients, peripheral nerve injury manifests itself as localized hypoesthesia (e.g., numbness) or, less commonly, as focal hyperesthesia along the greater occipital, lesser occipital, or auriculotemporal nerves. Hyperesthesia, in turn, may be exhibited in the form of hyperalgesia (normally painful stimuli become more painful) or allodynia (non-noxious sensations evoke pain) and may be referred to the superior aspect of the scalp.7
Figure 1. The Safe Donor Area (shaded zone) from which donor hair follicles can be safely harvested for subsequent transplantation overlaps the pathway of several sensory nerves: the greater and lesser occipital as well as the auriculotemporal nerves. Often these neuronal fibers are temporarily injured during classic strip donor removal.
Efforts to minimize or avoid these sensory sequelae in HRS include pre-excisional use of bilevel saline tumescence, limiting the use of electrocautery, intraoperative injection of corticosteroids and even local administration of Botulinum Toxin Type-A (BOTOX®; Allergan, Irvine, CA, USA).8 This study was conducted to determine which patient characteristics or intraoperative maneuvers commonly used within the investigated surgical practice influence occurrences of postoperative hyperesthesia.
Materials and Methods
This multicenter retrospective chart review involved 552 patients who underwent hair transplant surgery from 1999 to 2009 at one of four surgical centers: one private center in Toronto, ON, Canada, and three centers (two private surgical clinics and one 1171-bed hospital) in New York, NY, USA. The five surgeons participating in the study occasionally performed cases in more than one center, and their shared medical staff also assisted at various sites throughout the study. Patient charts were chosen at random from among those treated for the first time within the investigated practice since 1999.
The measured binary outcome was the presence/absence of patient-reported hyperalgesia or allodynia (not hypoesthesia) within the donor or recipient region of the scalp more than 1 month postoperatively which was bothersome enough to the patient that the physician felt it required treatment with direct injection of a triamcinolone (TMC) acetonide (Kenalog®; Bristol-Myers Squibb, Princeton, NJ, USA) solution targeting the effected nerve. Only patients who did not have donor hyperesthesia as a baseline prior to HRS were considered positive (n = 19).
Electrocautery usage determinations were based on nonretrospective observations of each individual surgeon’s hemostasis technique (reflecting the measured time unique to each surgeon’s technique). Work (kJ) performed for each technique was calculated from the amount of time the McKesson® (San Francisco, CA, USA) 22-900 High-Frequency Desiccator on a monopolar setting was used by the respective surgeon and multiplying that by power (watts). From this series of observations (n = 51), two distinct averages of usage emerged (one-way analysis of variance (anova): P < 0.0001): minimal usage = 0.572 kJ (surgeons A = 0.5599 [n = 14]; B = 0.6071 [n = 8]; and C = 0.549 [n = 7]) and moderate electrocautery = 1.983 kJ (surgeons D = 1.936 [n = 15] and E = 2.03 [n = 7]).
Measured intraoperative, bilevel sterile saline tumescence was administered via a 20G needle at both the intradermal level and the subfollicular plane within subcutaneous tissue (approximately 1–3 mm and 5–7 deep to the stratum corneum, respectively).
Tension measurements upon donor wound edge closure with sterile nylon nonabsorbable sutures (either 2-0 or 3-0 Supramid Extra® II; S. Jackson, Inc., Alexandria, VA, USA) were categorized in three levels: positive tension, zero (or “no”) tension, and negative tension. The tension categorization was based on the amount of scalp overlap possible when the superior and inferior edges of the donor wound are pulled together with forceps prior to closure [Fig. 2]. Positive tension is documented for wound edges that cannot be easily approximated with forceps and require additional manipulation (specifically, undermining of wound edges) to facilitate closure of the donor wound. If the edges can be brought together with forceps but no overlap is possible, the donor closure is considered to have zero tension. If overlap is achievable, negative tension is recorded. The presence or absence of intraoperative injection of Kenalog® (Bristol-Myers Squibb) 3.33 mg/mL (10 mg/mL diluted in 2% xylocaine with 1:200 000 epinephrine) was also investigated. This corticosteroid was applied in one of two manners: either directly into the donor wound in close proximity with observed large vascular bundles (in presumed tandem with nerve pathways) prior to wound closure or 2–3 mm inferior to the wound edge along the length of the donor area after it has been sutured closed.
Figure 2. Variations in tension during donor wound edge closure may lead to postoperative hyperesthesia via nerve compression. A simple scale to quantify scalp donor closing tension defines wound edges that cannot be easily brought together with forceps and require additional manipulation (i.e. undermining of would edges) to facilitate proper closure as “positive tension” (a); edges that can be easily approximated with forceps, but no overlap is achieved as “no tension” (b); and overlap is easily manageable as “negative tension” (c).
Data were analyzed using GraphPad InStat® software Version 3.1a (La Jolla, CA, USA). For measuring single independent variables (e.g., presence vs. absence of intraoperative tumescence, etc.), Fisher’s exact tests with two-tailed P values were utilized along with odds ratios and 95% confidence intervals using the approximation of Woolf. Variables defined by more than two categories (e.g., range of follicular units (FU) per surgery, etc.) were analyzed using a chi-squared test for trend with one degree of freedom. A chi-squared test for independence was used when more than two categories were not arranged with natural numerical spacing (e.g., tension). Two-tailed P values were subsequently calculated from these data. A P value was considered statistically significant.
Results and Discussion
A total of 19 of the 552 patients (3.4%) evaluated reported postoperative hyperesthesia. Only one variable reached statistical significance in influencing the frequency of this neurosensory symptom: patients who had undergone a previous hair transplant by a physician practicing outside of the five-physician medical group investigated in this study (n = 35) had more than a 14% incidence of this unusual postoperative complication after a session with one of the five investigated surgeons, while no patient who had opted for a procedure subsequent to surgeries previously performed exclusively within the investigated practice (n = 42) experienced this symptom (P = 0.0404) [Table 1]. There was no linear association, however, between the number of subsequent hair restoration surgeries and the incidence of hyperesthesia. None of the 19 patients reporting postoperative hyperesthesia experienced those neurologic symptoms prior to their surgery.Table 1. Analysis of trends in incidences of postoperative hyperesthesia in hair restoration surgery patients. Measured variables include variations in patient gender, age (years), the size of hair transplant session (number of follicular units (FU) transplanted), extent of intraoperative electrocautery, (kJ), use of bilevel saline tumescence prior to donor harvest, degree of tension upon donor closure, undermining prior to donor closure (cm), injection of intralesional triamcinolone acetonide, number of prior hair restoration surgery (HRS) sessions, and the effect of having had an initial hair transplant session at a practice different from that which performs subsequent surgeries. Only one variable, having undergone a previous hair transplant surgery outside of the surgical practice investigated in this study, significantly increased the frequency of postoperative hyperesthesia (n = 35, P = 0.0404)
|Hyperesthesia||Normal||Total||% Total||% Incidence|
|Gender (P = 0.7807)*|
|Age (P = 0.09544)†|
|Size (FUs) (P = 0.6420)‡|
|Electrocautery (P = 0.0897)§|
|Minimal (0.572 kJ)||3||188||191||35||1.57|
|Moderate (1.983 kJ)||16||345||361||65||4.43|
|Tumescence (P = 1.000)¶|
|Tension (P = 0.3044)**|
|Undermining (P = 0.4420)††|
|Undermine 1 cm||2||73||75||13||2.67|
|Undermine >1 cm||2||23||25||5||8.00|
|Intralesional triamcinolone injection (P = 0.7880)‡‡|
|Prior HRS (P = 0.3285)§§|
|Changing surgeon performing subsequent HRS (P = 0.0404)¶¶|
|Diff Surgeon within same practice||0||42||42||8||0.00|
|Diff Surgeon outside of Practice||5||30||35||6||14.29|
|Overall changes of Surgeons||5||72||77||14||6.49|
Although female patients reported a slightly higher incidence of this complication than men (3.97–3.28%, respectively), a significant association with gender did not emerge (P = 0.7807) [Table 1]. Age does not appear to be a factor influencing the rate of postoperative hyperalgesia or allodynia (P = 0.9544). While the peak age group experiencing this symptom was 41- to 50-yr-old patients (6.43%), the reported incidences declined as patients reached their 50s and beyond (2.63% and 2.44%, respectively). The size of the surgery can be measured in terms of FU transplanted. No trend appeared between the number of FU and this complication (P = 0.6420). Peak incidence reported by patients who had surgical sessions smaller than 1000 FU (6.25%) likely reflected the majority of patients with minimal scalp laxity (a requisite for reducing donor closure tension); it may be an additional indirect measure of donor tension or an accumulation of nerve trauma rather than a simple reflection of session size.
Delivery of high-frequency electrocautery damages both pain-conducting unmyelinated nerve fibers and myelin fibers conducting epicritic sensibility.9 While not reaching statistical significance, more than double the percentage of patients receiving moderate electrocautery (1.983 kJ) to maintain hemostasis along the donor wound reported postoperative hyperesthesia vs. their cohorts in which only a minimal amount of electrocautery (0.572 kJ) was applied (4.43% and 1.57%, respectively; P = 0.0897). Although use of bilevel tumescence prior to donor strip removal theoretically minimizes direct trauma to peripheral nerves by separating the nerve plexus from the subfollicular plane through which the scalpel passes, it did not alter the reported incidence of postoperative hyperesthesia (3.44%vs. 3.45% without tumescence).10 Furthermore, the quantity of tumescence used (ranging from 5 mL to 60 mL) did not influence this reported complication (P = 0.9269, not shown).
Traction along a neural pathway or compression of a nerve may also evoke action potentials in the nociceptive afferent fibers manifesting clinically as hyperesthesia.11–13 However, the observed association between the degree of donor closing tension and the frequency of postoperative hyperesthesia failed to achieve statistical significance (P = 0.3044) despite only 2.55% of patients reporting this complication after having undergone donor wound closure with “negative tension” as compared to 5.19% with no tension and 4.69% with “positive tension.” Undermining the donor wound edges prior to donor closure also led to more frequent reports of neuralgia (4% incidence vs. 3.32% without undermining) (P = 0.7612). Although also not reaching statistical significance, the extent of undermining influenced the reports of hyperesthesia, with nearly a fourfold increase in reported hyperesthesia as undermining requirements expanded from 1 cm to 2 cm beneath the donor wound edges (2.67–8.0%, respectively) (P = 0.4420). Use of Kenalog® (Bristol-Myers Squibb) was paradoxically associated with a slightly higher rate of postoperative hyperesthesia (3.76% incidence vs. 3.33% when not applied) (P = 0.7877). Interestingly, as the total dose of Kenalog® (Bristol-Myers Squibb) increased (from 10 mg to 20 mg to over 30 mg), a marked decrease in reported hyperesthesia was observed (5.55%, 3.19%, 0%, respectively) (P = 0.4720, not shown).
Of note, while many patients examined in this study had complicated past medical histories (HIV, diabetes, etc.), only one of the 19 patients reporting focal hyperalgesia or allodynia postoperatively had an irregular past medical history (osteoporosis). Another of the 19 patients requested that an anesthesiologist be present during the surgery owing to his “extremely low pain tolerance,” and two other patients reporting postoperative hyperesthesia had requested multiple preoperative consultations to allay their anxieties related to the surgery. It should also be mentioned that there were no indications of infection (skin induration, etc.) among the patients reporting postoperative hyperesthesia. In the absence of such suggestive external symptoms, the etiology of nerve injury was assumed and Kenalog® (Bristol-Myers Squibb) was injected.
Herein, we describe the first retrospective analysis of demographic and technical variables that may predispose HRS patients to postoperative hyperesthesia. This complication may be treated with local injection of Kenalog® (Bristol-Myers Squibb) (triamcinolone acetonide) 3.33 mg/mL in 2% xylocaine solution to the effected nerve. Typically, focal administration (1–3 mL) of this synthetic glucocorticosteroid with marked anti-inflammatory action is adequate. However, effective resolution may require two (or rarely three) treatments at 3-week intervals. This steroid is thought to decrease hyperalgesia or allodynia by two mechanisms: minimizing nerve inflammation and suppressing the nerve firing induced by prostaglandin E2, a chemical mediator that acts on peripheral nociceptors.14 Despite the tendency of sensory nerves to be less refractory than motor nerves to full recovery, these neurosensory symptoms generally decrease even without treatment over the course of 3–18 months based on age-related differences in the plasticity of the peripheral nervous system after injury.15,16
Although a number of variables were examined in this study, only one trend achieved statistical significance in influencing the rate of postoperative hyperesthesia: patients who had undergone a previous HRS by a physician not practicing within the same five-surgeon medical group investigated in this study experienced a markedly increased rate of this complication after a session within the investigated practice (P = 0.0404). Conversely, no patients undergoing a subsequent session within the investigated surgical group after having had all previous procedures within the same practice experienced these symptoms. These results suggest that minimizing the variation in surgical technique with subsequent procedures may help reduce the incidence of postoperative hyperesthesia and that physicians planning a repair of a patient who has undergone a prior transplant performed at a different surgical practice would be wise to include this caveat in their preoperative consultation.
Between patients initiating within the investigated surgical group and those having undergone HRS outside of the practice, variations in depth of scalpel incision may be the most likely contributor to this curious trend. Having trained under the same senior surgeon, each of the five physicians within the investigated practice share a common approach to donor harvesting (i.e., depth of incision, degree of undermining, and superficial suturing technique). This uniform approach was markedly altered in cases involving increased depth of donor scar tissue, a frequent challenge in patients having gone HRS elsewhere and seeking correction of their wider-than-average donor scar. An important component of this corrective process is the removal of excess scar tissue (which differs from undermining in that it takes place within the limits of the donor wound edges) that often requires deeper scalpel penetration through subcutaneous tissue where delicate nerve plexes abound. The difference may also stem from a nonsurgical approach to patient care: a variation may exist between practices regarding the degree to which patients were made aware of the correlation between this symptom and the surgery. Discussion of the possible development of postoperative hyperesthesia is a consistent emphasis of each preoperative consultation and postoperative follow-up in the investigated practice.
Interestingly, commonly accepted techniques used to avoid postoperative hyperesthesia (e.g., use of bilevel tumescence and intralesional Kenalog® [Bristol-Myers Squibb] injections) failed to reduce those incidences to a significant degree. While trends reflecting the influence of each factor began to emerge in this study, increasing the overall size of the patient population analyzed would demonstrate more unequivocally the influence of each variable on this sensory sequella. Furthermore, case-controlled analyses focused exclusively on each variable discussed may ultimately corroborate the theoretical benefit of a particular surgical maneuver or technique employed to combat this complication. Nevertheless, this analysis represents an important initiation in our effort to not only improve the esthetic outcome but also most effectively establish accurate expectations in our patients.
- Seery GE. Surgical anatomy of the scalp. Dermatol Surg 2002; 28: 581–7.
- Perez-Meza D, Niedbalski R. Complications in hair restoration surgery. Oral Maxillofac Surg Clin North Am 2009; 21(1):119–48.
- Seery GE. Scalp surgery: anatomic and biomechanical considerations. Dermatol Surg 2001; 27: 827–33.
- Rose P, Leavitt M. Anatomy of the scalp. In: WP Unger, R Shapiro, R Unger, M Unger, eds. Hair Transplantation, 5th edn. New York: Marcel Dekker; 2010: pp. 8–11.
- Leonard RT, Sideris K. Survey finds demand for hair restoration continues to grow: number of patients worldwide increased 26% since 2006. HT Forum Intrnl 2009; 19: 113–7.
- Murray BE. Encyclopedia of the Neurological Sciences: Nerve Injury. New York: Elsevier; 2003: pp. 418–21.
- Andrew D, Greenspan JD. Mechanical and heat sensitization of cutaneous nociceptors after peripheral inflammation in the rat. J Neurophysiol 1999; 82: 2649–56.
- Taylor M, Silva S, Cottrell C. Botulinum toxin Type-A (BOTOX®) in the treatment of occipital neuralgia: a pilot study. Headache 2008; 48: 1476–81.
- Fröhling MA, Schlote W, Wolburg-Buchholz K. Nonselective nerve fibre damage in peripheral nerves after experimental thermocoagulation. Acta Neurochir (Wien) 1998; 140: 1297–302.
- Field LM, Namias A. Bilevel tumescent anesthetic infiltration for hair transplantation. Dermatol Surg 1997; 23: 289–90.
- Weimer LH. Encyclopedia of the Neurological Sciences: Neuropathies, Entrapment. New York: Elsevier; 2003: pp. 535–40.
- McLachlan EM, Jänig W, Devor M. Peripheral nerve injury triggers noradrenergic sprouting within dorsal root ganglia. Nature 1993; 363: 543–6.
- Lui T, Knight KR, Tracey DJ. Hyperalgesia due to nerve injury – role of peroxynitrite. Neuroscience 2000; 97(1):125–31.
- Muramoto T, Atsuta Y, Iwahara T et al. The action of prostaglandin E2 and triamcinolone acetonide on the firing activity of lumbar nerve roots. Int Orthop 1997; 21: 172–5.
- Evans GR. A review and approach to tissue engineered constructs. Anat Rec 2001; 263: 396–404.
- Kovacic U, Sketelj J, Bajrovi? FF. Age-related differences in the reinnervation after peripheral nerve injury. Int Rev Neurobiol 2009; 87: 465–82.
To share ideas with other hair loss sufferers visit the Hair Restoration Social Network and Enhanced Discussion Forum
Get Proven Treatments at the Best Prices by visiting our new Online Hair Loss Treatment Shop.