Relative Effects of Minoxidil 5%, Platelet-Rich Plasma, and Microneedling in Pattern Hair Loss: A Systematic Review and Network Meta-Analysis

Abstract

Background

Combination treatments may improve the utility of approved agents for the treatment of pattern hair loss (PHL); however, head-to-head comparisons are lacking.

Objective

The aim of the study was to compare the efficacy of 5% minoxidil, platelet-rich plasma (PRP), and microneedling across adults with PHL insofar as change in total hair density at 24 weeks.

Methods

We conducted a literature search in July 2022. Through our Bayesian network meta-analysis, we estimated treatments’ surface under the cumulative ranking distribution (SUCRA) values and relative effects – in terms of mean difference (MD).

Results

Data from 27 trials, totaling 1,110 patients, were extracted. Interventions were ranked based on the probability of inducing hair density improvements: 5% minoxidil plus microneedling (SUCRA = 95.8%), 5% minoxidil plus PRP (SUCRA = 64.7%), 5% minoxidil (SUCRA = 53.9%), PRP (SUCRA = 34.9%), microneedling (SUCRA = 27.8%), and PRP with microneedling (SUCRA = 22.9%). The efficacy of 5% minoxidil plus microneedling in improving total hair density was significantly greater (p < 0.05) than 5% minoxidil monotherapy (MD = 13 hairs/cm²), PRP monotherapy (MD = 16 hairs/cm²), and microneedling monotherapy (MD = 17 hairs/cm²).

Conclusion

Five percent minoxidil plus microneedling is an effective treatment option for improving hair density at 6 months in adult PHL patients.

Introduction

The connection between hair loss and psychological distress is evident throughout the scientific literature [1, 2]; androgenetic alopecia (AGA) – which is also referred to as pattern hair loss (PHL) – is the most common form of hair loss among men and women [3]. While the knowledge of the condition’s etiology is far from complete, it has been well established that AGA is linked to aberrations in androgen metabolism [3].

Systemic therapeutic agents for AGA include 5-alpha reductase inhibitors (such as finasteride and dutasteride) and low-dose oral minoxidil [4]. Nonsystemic treatment modalities include topical solution of 5% minoxidil, platelet-rich plasma (PRP), and microneedling [5]. Microneedling involves puncturing patients’ scalp regions with medical-grade needles – that are less than 5 mm in depth to obtain cosmetic improvement in hair count [5]. In the current study, we conducted a network meta-analysis (NMA) to determine the relative effects of mono- and poly-treatment with 5% minoxidil, PRP, and microneedling in improving the total hair density for adults with PHL at 6 months.

Materials and Methods

The protocol for our work was published in the International Platform of Registered Systematic Review and Meta-Analysis Protocols (INPLASY) database under the ID: 202290042; our work also followed the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines [6].

Search for Eligible Studies

As per the patient (P), intervention (I), comparator (C), and outcome (O) framework for our study’s objective, we determined the relative efficacy of 5% minoxidil, PRP, and microneedling as stand-alone therapies or various combinations thereof (I/C) on adults (i.e., aged at least 18 years) with PHL (P) in terms of change in total hair density at 24 weeks (O). We chose this outcome as it is not only a common endpoint across AGA trials but is also a clinically relevant one [4, 7].

The data for our quantitative analysis were obtained from studies that were identified through a systematic search of the peer-reviewed literature. The search was conducted using PubMed and Scopus on July 24, 2022, without date restrictions. Items identified using the search term “micro*” and “needl*” were combined the results obtained from the following MeSH/search terms: “alopecia,” “androgenetic alopecia,” “baldness,” “hair loss,” “platelet rich plasma,” “minoxidil.” We also mined the bibliographies of relevant articles. Non-English articles were excluded. The literature search, title/abstract screen, and full-text review were performed independently by two authors (S.P.R. and T.W.); discrepancies at any stage were resolved through discussion with a third author (A.K.G.). We used Rayyan for the coordination of our systematic review [8]. Our parameters for data extraction include authorship, year of publication, interventions’ details, sample size, patients’ characteristics (sex and disease severity), and design of treatment administration (whole-head vs. split-scalp); these variables were chosen a priori as findings from previous studies support that they can influence the efficacy of AGA therapies [7, 9].

Statistical Analyses

Direct comparisons between interventions across trials were depicted with a network plot, a diagram of nodes and edges. A node corresponds to a given intervention; an edge is represented by a line between two nodes and corresponds to the direct comparison of two interventions from a head-to-head trial [10].

In our NMA, we adjusted for patients’ sex and design of study (i.e., whole-head vs. split-scalp); hence, our NMA was a network meta-regression, a form of NMA that simultaneously determines the relative impact of interventions while adjusting for variation due to effect modifiers [11]. We statistically adjusted for the effect of sex because dihydrotestosterone is implicated in PHL; moreover, efficacy of agents in AGA trials tends to be studied separately in men and women. Our adjustment for study design is based on the previous studies’ findings which support that PRP’s efficacy can be substantially influenced by whether the design of the trial was of a split-scalp or whole-head design [7]. Furthermore, we selected uniform priors for our Bayesian analyses – where we used 10,000 iterations, 1,000 burn-ins; Bayesian analyses were based on Markov chain Monte Carlo sampling. Results from the random-effects model were chosen over those of the fixed-effects as the former had better fit (as per deviance information criterion values). Our statistical analysis was conducted with RStudio [12]; the BUGsnet and multinma packages were used [11, 13]. We ran a Bayesian multivariable network meta-regression – from which interventions’ surface under the cumulative ranking curve (SUCRA) values, and relative effects (as per mean difference), were estimated. A treatment’s SUCRA value corresponds to its overall rank for efficacy; a relative effect represents the comparative efficacy between two interventions. For our study, the relative effect of every pairwise comparison was quantified, and the respective 95% credible interval thereof were also estimated. For our analysis, alpha (i.e., significance level) was set to 5%. We used the updated Cochrane Collaboration’s risk of bias 2 tool, to assess evidence quality for the individual studies [14].

Interpretation of a treatment’s ranking for efficacy is more comprehensive when statistical findings and assessment of evidence quality are conjunctly considered; conclusions are less valid if they are solely based on SUCRA values. Hence, wherever relevant, we reported relative effects, quality assessments, and SUCRA values together.

Results

Included Studies and Network Characteristics

Twenty-seven studies met our eligibility criteria for quantitative analysis (Fig. 1) [15–41]. The PRISMA checklist is presented in online supplementary eTable 1 (for all online suppl. material, see https://doi.org/10.1159/000534196). There were a total of 1,110 patients, and 36 arms of interventions – across which a total of six types of interventions were compared, namely, topical minoxidil 5% and microneedling (3 arms), topical minoxidil 5% alone (15 arms), PRP alone (12 arms), PRP with 5% minoxidil (1 arm), microneedling alone (3 arms), and microneedling with PRP (2 arms). Arm-level study characteristics are presented in Table 1, and Figure 2 illustrates the qualitative evaluation of each trial’s risk of bias. The majority of the arms had only male subjects (23/27). A network plot of the six interventions is depicted in Figure 3.

Fig. 1.

Schematic for identification of studies.

Table 1.

Study characteristics of 27 trials whose data were used in analyses

Study	Treatment	Regimen	Sample size, na	Proportion of male subjects	Age, mean (SD)	Disease severityb	Design (whole-scalp or split-scalp)
Aggarwal et al. [15], 2020	Microneedling	4 monthly sessions	30	1	25.50 (5.35)	Grade III = 70%	Split
	Microneedling + PRP
						Grade II = 30%
Alves and Grimalt [16], 2016	PRP	3 monthly sessions	25	0.48	39.0 (range; 21.0–62.0)	Male: grades II–V	Split
						Female: grades I–III
Bao et al. [17], 2020	Microneedling	12 sessions, 2 weeks apart	18	1	34.1 (3.1)	Grades III, IV = 56%	Whole
						Grades V, VI = 44%
	Minoxidil 5%	1 mL BID	18	1	34.7 (6.9)	Grades III, IV = 73%	Whole
						Grades V, VI = 27%
	Microneedling + minoxidil 5%	12 sessions, 2 weeks apart	20	1	35.2 (3.3)	Grades III, IV = 55%	Whole
		1 mL of minoxidil 5% BIDc				Grades V, VI = 45%
Bao et al. [18], 2022	Microneedling	8 sessions, 3 weeks apart	23	1	38.4 (8.5)	Grades III, IV = 57%	Whole
						Grades V, VI = 43%
	Minoxidil 5%	1 mL BID	23	1	37.0 (8.4)	Grades III, IV = 57%	Whole
						Grades V, VI = 43%
	Microneedling + minoxidil 5%	8 sessions, 3 weeks apart	25	1	36.3 (8.0)	Grades III, IV = 52%	Whole
		1 mL of minoxidil 5% BIDc				Grades V, VI = 48%
Berger et al. [19], 2003	Minoxidil 5%	1 mL BID	41	1	40.0 (range; 18.0–49.0)	Grades III–IV	Whole
Butt et al. [20], 2019	PRP	2 monthly sessions	30	0.67	28.7 (7.7)	Male: grades III–VI	Whole
						Female: grades I–III
Chuah et al. [21], 2023	PRP	4 sessions, 3 weeks apart	50	0.62	38.7 (10.5)	Male: grade III–VI	Split
						Female: grades II–III
Dicle et al. [22], 2020	PRP	3 monthly sessions	15	1	31.9 (8.8)	Grades III, IV = 53%	Whole
						Grade V = 27%
Faghihi et al. [23], 2022	Minoxidil 5%	1 mL BID	20	0.55	27.1 (4.8)	Male: grades II–V	Whole
						Female: grades I–II
Ferrara et al. [24], 2021	Minoxidil 5%	1 mL BID	19	1	41.7 (6.8)	Grades III–VI	Split
Gupta et al. [25], 2017	Microneedling + PRP	6 sessions, 15 days apart	30	1	28.3 (3.1)	Grades III, IV = 30%	Whole
						Grades V, VI = 70%
Hausauer and Jones [26], 2018	PRP	2 sessions, 3 months apart	19	0.63	46.9	Male: grades II–V	Whole
						Female: grades I–II
	PRP	3 monthly sessions plus booster session at 6 months	20	0.85	40.1	Male: grades II–V	Whole
						Female: grades I–II
Kapoor et al. [27], 2020	PRP	8 sessions, 3 weeks apart	25	1	Range; 25.0–50.0	Grade II = 32%	Whole
						Grades III, IV = 68%
Kozicka et al. [28], 2022	Minoxidil 5%	QD	76	0	49.0	NR	Whole
Liang et al. [29], 2022	Minoxidil 5%	1 mL QD	38	0	31.1 (6.9)	Sinclair grades II–III	Whole
	Microneedling + minoxidil 5%	12 sessions, 2 weeks apart	40	0	30.8 (6.3)	Sinclair grades II–III	Whole
		1 mL of minoxidil 5% QD
Hossein Mostafa et al. [30], 2021	Minoxidil 5%	1 mL BID	12	1	33.2 (8.4)	Grade II = 42%	Whole
						Grades III, IV = 58%
Pachar et al. [31], 2022	Minoxidil 5%	1 mL BID	44	1	18.0–54.0	Grade II = 42%	Split
						Grades III, IV = 46%
						Grade V = 12%
	PRP + minoxidil 5%	7 monthly sessions	44	1	18.0–54.0	Grade II = 42%	Split
		1 mL of minoxidil 5% BID				Grades III, IV = 46%
						Grade V = 12%
Pumthong et al. [32], 2012	Minoxidil 5%	BID	22	1	38.0 (range; 21.0–57.0)	Grade II = 9%	Whole
						Grades III, IV = 45%
						Grade V–VII = 45%
Qu et al. [33], 2019	PRP	6 monthly sessions	93	0.55	Male: grade II – 26.8; grade III – 27.5; grade IV – 28.6; grade V – 30.5	Male: grades II (24%), III–IV (47%), V (29%)	Whole
					Female: grade I – 27.6; grade II – 28.5; grade III – 29.9	Female: grades I (31%), II (36%), III (33%)
Qu et al. [34], 2021	PRP	3 monthly sessions	52	0.62	NR	Male: grades II–V	Split
						Female: grades I–III
Sakr et al. [35], 2013	Minoxidil 5%	1 mL BID	11	1	Range; 25.0–30.0	Grade II–IV	Whole
Salah et al. [36], 2020	Minoxidil 5%	1 mL BID	15	1	31.5 (6.6)	Grade II = 7%	Whole
						Grades III, IV = 64%
						Grades V, VI = 29%
Sasaki [37], 2021	PRP	2 sessions, 3 months apart	8	0.5	Male: 51.8 (12.7)	Male: grades III–IV	Split
						Female: grades I–II
	PRP	2 sessions, 3 months apart	8	0.5	Female: 51.3 (13.9)	Male: grades III–IV	Split
						Female: grades I–II
Suchonwanit et al. [38], 2019	Minoxidil 5%	1 mL BID	29	1	35.4 (10.3)	Grades III, IV = 57%	Split
						Grades V, VI = 43%
Tan et al. [39], 2021	Minoxidil 5%	BID	16	1	30.8 (6.6)	BASP grades M1–M3, C1–C2, V1–V3, F1–F3	Split
Tsuboi et al. [40], 2009	Minoxidil 5%	1 mL BID	141	1	40.7 (6.9)	Ogata grade II	Whole
Zhou et al. [41], 2020	PRP	3 monthly sessions	10	1	41.6 (4.1)	Class III, IV = 90%	Split
						Class V = 10%

BID, twice a day; NR, not reported; PRP, platelet-rich plasma; QD, once daily.

^aNumber of patients at follow-up.

^bThe Hamilton-Norwood scale was applied to male subjects; the Ludwig scale was applied to female subjects.

^c2 mL of minoxidil 5% was applied on the day of microneedling session.

Fig. 2.

Risk of bias was assessed for the 27 studies that were eligible for quantitative analysis across five domains. The overall risk of bias for each study is also presented. For each domain, the studies received one of three ratings: namely, low risk of bias (denoted by the plus sign in green circle), some concerns (denoted by the question mark in yellow circle), or high risk of bias (denoted by the minus sign in red circle).

Fig. 3.

Network plot for the agents whose efficacy was simultaneously compared in our NMA.

Relative Efficacy of Mono- and Poly-Therapy with Microneedling

Across the 36 arms, there was complete data for patients’ sex and study design; so, these two covariates were adjusted for in our network meta-regression. Interventions’ SUCRA values and relative effects are presented in Tables 2 and 3, respectively. Microneedling with 5% minoxidil was ranked the most effective of the six interventions (SUCRA = 96%); the 6-month increase in total hair density with the highest ranked treatment modality was significantly (p < 0.05) greater than monotreatment with: topical minoxidil 5% (mean difference (hairs/cm²) = 13.01, 95% CI: 1.66–24.01), PRP (mean difference (hairs/cm²) = 16, 95% CI: 2.57–28.66), and microneedling alone (mean difference (hairs/cm²) = 17, 95% CI: 3.12–30.54) (Table 3).

Table 2.

Agents’ surface under the cumulative ranking curve (SUCRA) values for network meta-analysis (NMA) on efficacy as per change in total hair density at 24 weeks for patients with PHL

Regimen	SUCRA, %
Microneedling and 5% minoxidil	95.81
PRP and 5% minoxidil	64.68
5% minoxidil	53.925
PRP	34.905
Microneedling	27.78
Microneedling and PRP	22.9

The NMA adjusted for variation to biological sex (i.e., male vs. female) and design of study (i.e., whole-head vs. split-scalp).

Table 3.

Relative effects – as per mean difference – of agents whose efficacy were compared in pairwise manner

Monotherapy with PRP (SUCRA = 34.9%) was ranked lower than solo treatment with topical minoxidil 5% (SUCRA = 53.9%); however, there was no statistical difference in their effects (p > 0.05). Similarly, the effect of monotherapy with microneedling (SUCRA = 27.8%) and PRP (SUCRA = 34.9%) were not statistically different (p > 0.05) (Table 3).

Our work has strengths and limitations. Given that Jenkins and Quintana-Ascencio [42] pointed out the tendency for meta-regression to be of small sample size (i.e., less than 25 studies), our study did not have this limitation as we had a total of 36 arms – across which there were 1,137 patients. As Thompson and Higgins [43] pointed out, a common limitation of meta-regression studies is that the findings thereof could be subject to the “ecological fallacy,” a phenomenon that is also referred to as “aggregation bias”. An ecological fallacy occurs when an association that is observed at the aggregate level is nonexistent at the individual level. Hence, our findings make a case for the conduct of multi-arm AGA trials on mono- and poly-therapy with microneedling from which individual-level data analyses can be made. Conclusions from our aggregate analyses may be congruent– and thereby strengthen – patient-level results.

Discussion

In practice, it is not always feasible to conduct head-to-head AGA trials with all possible comparators; however, the NMA technique allows for the simultaneous comparison of various interventions’ therapeutic effect. The data for this analysis were obtained through a systematic review of the relevant peer-reviewed literature. Through employing the technique of multivariable network meta-regression, an analysis of the relative effects of three or more interventions was simultaneously determined while adjusting for variation of confounding variables [11].

A strength of our study design is that the outcome data across each of the 36 arms pertained to the same time point (i.e., 24 weeks). This design, therefore, eliminates selection bias that can result from analyzing an outcome across different time points (i.e., time-varying confounding).

Our findings are congruent with the literature; for example, the results of study resonate with the meta-analysis study by Abdi and colleagues [44]. The authors pooled data across 8 trials to compare the effect of microneedling and topical minoxidil with that of topical minoxidil alone on change in total hair density. Like our study, the authors also found that combination therapy (i.e., topical minoxidil and microneedling) is significantly (p < 0.00001) more effective than topical minoxidil alone [44].

Our NMA showed that, on average, topical minoxidil 5% and microneedling (SUCRA = 95.8%) was significantly (p < 0.05) more effective (as per 6-month change in total hair density) than minoxidil (5%) alone (SUCRA = 53.9%), PRP alone (SUCRA = 34.9%), or microneedling alone (SUCRA = 27.8%) and after covariate adjustment. Given that the literature has growing evidence of the therapeutic effects of topical finasteride, a future extension of our work could include comparing the therapeutic effect of microneedling and topical finasteride or microneedling and topical dutasteride with those of the combination therapies investigated herein.

Conclusions

We have produced statistical evidence on the comparative effectiveness of mono- and poly-therapy with minoxidil 5%, PRP, and microneedling. With the aid of advanced statistical techniques, our quantitative findings accounted for potential confounding due to established effect modifiers. To date, the cosmetic impact of the interventions we compared has not been directly compared in an actual trial [45]. Therefore, to our knowledge, this is the first NMA that compares the relative efficacies of monotherapy with minoxidil 5%, PRP, and microneedling and poly-therapy with these combinations in terms of change in total hair density after half a year.

Statement of Ethics

An ethics statement is not applicable because this study is based exclusively on published literature.

Conflict of Interest Statement

The authors have no conflict of interest to declare.

Funding Sources

No funding was received for this work.

Author Contributions

Conceptualization, A.K.G.; data curation, M.A.B., S.P.R., and T.W.; formal analysis and software, M.A.B.; investigation, S.P.R., T.W., M.T., and M.A.B.; methodology, M.A.B. and M.T.; project administration and resources, A.K.G.; supervision and validation, A.K.G. and M.A.B.; visualization, M.A.B. and T.W.; writing – original draft preparation, A.K.G., M.A.B., and S.P.R.; and writing – review and editing, A.K.G. and T.W.

Funding Statement

No funding was received for this work.

Data Availability Statement

All data generated or analyzed during this study are included in this article. Further inquiries can be directed to the corresponding author.