The Role of Lasers and Light Devices For The Treatment of Melasma
The Role of Lasers and Light Devices For The Treatment of Melasma
The Role of Lasers and Light Devices For The Treatment of Melasma
Chee Leok Goh
16.1 Introduction
C.L. Goh
National Skin Centre, Singapore, YLL School of Medicine, National University of Singapore,
1 Mandalay Road, Singapore 308205, Singapore
e-mail: [email protected]
Fig. 16.1 PIH 2 months following QS Nd:YAG 1064-nm laser using convention treatment proto-
col for treating melasma. This patient was treated with QS Nd:YAG 1064 nm 4 ns pulsed duration
and fluence of 8 J/cm2 (Photograph courtesy of National Skin Centre, Singapore)
The Q-switched (QS) lasers are lasers with short pulse duration in nanosecond
domain. They are suited to treat pigmentary skin disorders because of melanin’s
short thermal relaxation time and absorption spectrum. These lasers include the QS
ruby, alexandrite, and Nd:YAG lasers. The QS lasers are wonderful devices for
removing tattoos, nevus of Ota, Hori’s nevus, lentigines, and some other pigmentary
disorders. Initially, it was thought that adopting the same treatment protocol used to
treat other pigmentary skin disorders will also be suitable for melasma [2, 3]. But
experience showed that they are not, as severe PIH is common and often no improve-
ment could be achieved.
Taylor reported eight patients with melasma or PIH treated with the QS ruby laser
(694 nm, 40 ns pulse duration) at fluences of 15–7.5 J/cm2. It was shown that regard-
less of fluence, there was no permanent improvement of their melasma; in some cases,
darkening of melasma was even observed. Histologic sections of biopsy specimens
taken before and after treatment showed extracellular melanin immediately after the
procedure. Several months after the last treatment, epidermal pigmentation was back
to baseline levels, and dermal melanophages were focally increased [4].
Given these results, the Q-switched lasers used in conventional treatment param-
eter for pigmentary skin lesions are not a recommended method to treat melasma.
16.2.2 A
blative Skin Resurfacing (Erbium:
Yttrium-Aluminum-Garnet Lasers)
The erbium:YAG laser (Continuum Biomedical, Dublin, CA) emits 2940-nm laser
wavelengths. This wavelength is highly absorbed by tissue water as its chromophore
and is an effective ablative resurfacing wavelength. Manaloto et al. treated ten
16 The Role of Lasers and Light Devices for the Treatment of Melasma 145
patients of Fitzpatrick skin phototypes II–V with refractory melasma using the
erbium:YAG laser. Utilizing MASI scores and spectrophotometer for assessment,
the authors found that there was improvement immediately after the laser treatment.
However, all patients developed PIH after 3–6 weeks follow-up, despite prophylaxis
oral steroids for 5 days post procedure [5].
While the PIH improved with serial glycolic acid peels, this side effect appears
to outweigh the benefit derived from this ablative procedure.
16.2.3 C
ombination of Carbon Dioxide and QS
Alexandrite Laser
a b
Fig. 16.2 Significant improvement of melasma after three treatments at monthly interval with
non-ablative fractional laser. But recurrence appeared 3 months after stopping treatment
(Photographs courtesy of National Skin Centre, Singapore)
Fractional laser resurfacing is a procedure that uses laser light to create scattered
microzones of thermal damage on the skin. It does not cause confluent full-thickness
skin wounds, and the in-between normal undamaged skin acts as a reservoir to
regenerate the laser damaged skin. This results in more rapid skin repair and recov-
ery [8]. This laser is approved by the Food and Drug Administration (FDA) for the
treatment of melasma, periorbital rhytides, pigmented lesions, skin resurfacing,
acne scars, and surgical scars [9]. The microthermal zones of injury limit the area of
the skin that is damaged with each treatment, which may decrease the risk of PIH.
The transepidermal elimination of debris and tissue through the microthermal
treatment zones after injury could serve as an effective method of removing dermal
melanophages [10]. Good clinical results may be seen in some patients (Fig. 16.2).
Roshkar evaluated ten patients with Fitzpatrick skin phototypes III–V treated
with a fractional laser (Fraxel; Reliant Technologies, Palo Alto, CA) for four to six
sessions, 1–2 weeks apart [11]. None of the patients were pretreated with hydroqui-
none. The authors found that six out of the ten patients had 75–100 % clearing of
melasma based on clinical evaluation and 30 % had less than 25 % improvement.
The nonresponders were all Hispanic patients. There was a 10 % risk of
post-inflammatory hypopigmentation. The risk of delayed PIH and relapse were not
assessed (level of evidence, II-iii).
However, the incidence of post-inflammatory pigmentation (Fig. 16.3) following
fractional laser resurfacing in Asians is high ranging from 10 to 90 % [12]. It is
likely that those with melasma will experience higher risk of PIH. Hence such
modality of treatment is not suitable for Asian melasma patients especially those
with darker skin type.
Another study looked at the histopathologic effects of fractional laser technology
on melasma. The study failed to support the efficaciousness of fractional laser resur-
facing for melasma [13]. The authors treated ten patients with epidermal melasma
16 The Role of Lasers and Light Devices for the Treatment of Melasma 147
a b
Fig. 16.3 PIH following a single treatment with the non-ablative fractional laser treatment for
melasma (Photographs courtesy of National Skin Centre, Singapore)
who had Fitzpatrick skin phototypes III–IV every 2 weeks for four sessions. Biopsy
specimens were obtained before treatment and 3 months after the final treatment.
Sunscreen use was advocated but depigmenting agents were avoided. After treat-
ment, lesional skin showed a decrease in the number of epidermal melanocytes and
fewer enlarged melanocytes on electron microscopy; however, there was no correla-
tion between histologic improvement and investigator-rated improvement.
Several later studies also reported lack of efficaciousness of fractional lasers for
melasma. Lee et al. reported 25 melasma patients who received four monthly frac-
tional laser treatments who achieved reduction of mean MASI score from 7.6 to 6.2.
There was 60 % improvement at 4 weeks after treatment which deteriorated to 52 %
at 24 weeks after treatment. Mean melanin index decreased significantly after the
first two sessions, but it relapsed in subsequent follow-ups. The treatment did not
alter skin elasticity. Hyperpigmentation was observed in three of 23 subjects (13 %).
The authors concluded that fractional laser treatment of melasma led to some clini-
cal improvements, but it was not as efficacious as previously reported. They recom-
mend judicious use of fractional laser for melasma in Asian skin because of its
limited efficacy. There is also risk of PIH [14].
Wind et al. treated 29 melasma patients with a split-face study using four to five treat-
ments with non-ablative 1550-nm fractional laser on half the face compared to daily
topical triple combination therapy (hydroquinone 5 %, tretinoin 0.05 %, triamcinolone
acetonide 0.1 % cream) alone on the other half in a 15-week study. After the last treat-
ment session, patients were asked to apply the triple combination cream twice weekly
on both sides of the face during follow-up. Mean patient global assessment and satisfac-
tion were significantly lower at the fractional laser-treated side (p < 0.001). Physician
global assessment, melanin index, and L-value showed a significant worsening of hyper-
pigmentation at the fractional laser-treated side. At the triple cream-treated side, no sig-
nificant change was observed. At 6-month follow-up, most patients preferred the triple
cream treatment. Side effects of the fractional laser-treated side were erythema, burning
sensation, edema, and pain. Nine patients (31 %) developed PIH after two or more laser
sessions. The authors concluded that given the high rate of PIH, non-ablative 1550-nm
fractional laser at 15 mJ/microbeam is not recommendable in the treatment of melasma.
Triple cream treatment remains the gold standard treatment [15].
148 C.L. Goh
a b
Fig. 16.4 Good improvement of melasma following eight treatments of laser toning with the QS
Nd:YAG 1064-nm laser performed every 2 weeks. Fluence was 2.5 J/cm2, 6 mm spot size, 5 ns
pulsed duration. Recurrence appeared 3 months after cessation of treatment (Photographs courtesy
of National Skin Centre, Singapore)
Fig. 16.5 Guttate
hypomelanotic macules
following too frequent and
high-dose laser toning with
the QS Nd:YAG 1064-nm
laser for melasma
(Photograph courtesy of
National Skin Centre,
Singapore)
a b
Fig. 16.6 Fairly good response of melasma to six sessions at monthly interval of IPL
Intense pulsed light (IPL) is a broadband light source and is not a laser device. The
IPL emits light of wavelength stretching from 500 to 1200 nm. This spectrum of
wavelength falls within the absorption spectrum of melanin and oxyhemoglobin.
Hence, it can be used to treat superficial pigmentary lesions including melasma.
“Cutoff” filters can be placed across IPL sources to eliminate unwanted shorter
wavelengths to prevent epidermal burn. These filters are especially useful when
using IPL on darker skin type.
Several studies have reported improvements of melasma with IPL with minimal
side effects. But patients often require multiple treatments (Fig. 16.6).
A report from Taiwan compared IPL treatment with HQ against HQ alone on 33
patients with Fitzpatrick skin phototypes III and IV having mixed melasma.
Patients received four monthly sessions of IPL. Using the mexameter, the authors
calculated a relative melanin index (defined as the difference between the melanin
index of lesional skin and the melanin index of normal skin). In the IPL/HQ group
(17 females), 35 % had greater than 50 % improvement compared to the HQ-alone
group (16 control) with only 14 % experiencing greater than 50 % improvement.
The patients who received IPL treatment had a 39.8 % decrease in the relative
melanin index after four treatments (16 weeks), while the control group receiving
topical therapy alone had only an 11.6 % decrease. At 24-week posttreatment, the
improvement on the IPL-treated side had decreased less to a mean of 24.2 %, sug-
gesting the need for maintenance treatments. Side effects of the IPL included some
crusting lasting for 1–2 weeks; transient PIH seen in 12 % was resolved with the
use of HQ [26].
In another report from China, 89 Chinese females with predominantly mixed
melasma unresponsive to topical therapy and chemical peels were treated with IPL
every 3 weeks for four sessions. Patients were instructed to use broad-spectrum
sunscreen and avoid bleaching creams. Improvement of melasma was assessed
using the MASI score. Pigmentation and erythema were objectively measured with
a mexameter. Mean MASI scores dropped significantly, from 15.2 to 5.2 after four
sessions and to 4.5 at the 3-month follow-up visit. Epidermal melasma responded
16 The Role of Lasers and Light Devices for the Treatment of Melasma 151
better than the mixed type. The melanin index as measured by the mexameter
dropped from a mean value of 140.8 to a value of 119; the erythema index dropped
significantly as well. The most common side effects included temporary erythema
and edema and microcrusting. PIH was observed in three cases [27].
16.2.7 N
ewer Intense Pulsed Light Treatment Protocol
and Devices
16.2.7.1 Low-Fluence and Short-Pulse Intense Pulsed Light
Low-fluence and short-pulse IPL has been reported to be effective and safe in the
treatment of melasma. The low-fluence IPL has nanosecond-level pulse duration
and allows selective thermolysis of melanosomes with minimal side effects.
Twenty Korean adults with melasma were randomly assigned to two groups and
treated at fluences of 10 or 13 J/cm2 of IPL weekly over 6 weeks. Subjects were
evaluated at baseline and weekly during the 6 weeks of treatment and at 3 weeks
following the final treatment. Melanin and erythema indices were scored using a
spectrophotometer. The mMASI score of 20 patients at inclusion was 11.6. Both
10 J and 13 J IPL treatment groups had decreased mMASI scores from 2 weeks
onward at statistically significant levels. Both 10 J and 13 J IPL treatment groups
showed decreased melanin indices with statistically significant differences from 3
weeks onward. The authors concluded that a low-fluence IPL protocol could p rovide
effective treatment for melasma with minimal side effects in Asian skin [28].
A recent report on a prospective, split-face, randomized study of the efficacy and safety
of a novel fractionated IPL with microsecond domain to treat melasma in 30 Asian
women showed significant improvement of melasma. In this 14-week split-face study,
one side of the face received weekly fractionated IPL and the other side biweekly con-
ventional IPL. The non-inferiority of a weekly fractionated IPL regimen to a biweekly
conventional IPL regimen was verified by a lower margin of the 95 % confidence inter-
val for the difference in the MASI change from baseline of 2.61 for each side. This
value was greater than the previously determined non-inferiority margin of –2.68
(p < 0.025). On the fractionated IPL side, the mMASI score decreased continuously,
but in the conventional IPL group, the mMASI score rebounded during the treatment
course. The authors concluded that fractionated IPL shows moderate efficacy as a
melasma treatment and is therefore a good alternative to conventional IPL. Fractionated
IPL can also be used as a maintenance treatment for melasma [29].
IPL appears to be moderately effective in the treatment of melasma. It does not
clear nor cure the condition. However, because of its safety and a lower percentage
of producing PIH, IPL can be considered as an adjunct in the treatment of recalci-
trant melasma.
152 C.L. Goh
“Laser toning” with Q-switched Nd:YAG 1064-nm laser combined with IPL for the
treatment of melasma has been reported to enhance the efficaciousness of the two
individual procedures. Skin toning with the QS Nd:YAG laser targets deeper pig-
ment, while IPL targets a wide range of superficial cutaneous structures.
In a study among 20 females with mixed-type melasma on both cheeks, laser
toning with the QS Nd:YAG laser was done full face for five sessions at 1-week
interval. One side of the face was randomly assigned to receive additional three ses-
sions of IPL treatments at 2-week intervals. At 12 weeks after the last treatment, 18
patients completing the study showed both sides of the face with significant
improvement in their mMASI score and melanin indices. A more rapid improve-
ment of mMASI score and melanin indices, however, was observed on the com-
bined side. At the end of treatment, 55 % improvement and 37 % improvement of
melanin indices were observed on the combined side and monotherapy side, respec-
tively. The overall patients’ satisfaction was in favor of the combined side.
Recurrence occurred on both sides but there was still a significant decrease com-
pared to baseline. No serious side effect was noted. The authors concluded that
combination of “skin toning” with QS Nd:YAG lasers and IPL results in faster
clearance of melasma and is more effective than skin toning with QS Nd:YAG laser
alone. However, recurrence is still inevitable [30].
The vascular lasers were used to treat melasma with variable results. There are not
enough good scientific reports to confirm the role of vascular laser and copper bro-
mide lasers for the treatment of melasma at present.
The basis for the role of vasculature in melasma was reported by Kim et al. It was
shown that melasma lesions have, in addition to increased pigmentation, more elasto-
sis and vascularization than perilesional skin. Chromometer measurements were sig-
nificantly higher in the melasma lesion than the nonlesional skin. Histology showed
that factor VIIIa-related antigen staining showed a significant increase in the number
and size of dermal blood vessels in the lesional skin. Added to the significant relation-
ship between the number of vessels and pigmentation, expression of vascular endo-
thelial growth factor (VEGF) was significantly increased in melasma skin [31, 32].
Pulsed dye laser was reported to improve the topical treatment outcome of melasma
and prolong remission period. Passeron et al. carried out a controlled, randomized,
single-blind, split-face clinical trial evaluating the effectivity of dual treatment of fixed
triple combination cream (TCC) and PDL in the treatment of melasma. Patient
16 The Role of Lasers and Light Devices for the Treatment of Melasma 153
satisfaction was significantly greater for the combination treatment. Half of the patients
with dark skin type IV developed PIH. PDL in association with a bleaching cream
appears beneficial in treating melasma in patients with skin phototypes II and III [33].
Following the same study, Passeron reported that melasma lesion treated with
PDL and triple cream showed long remission indicating enhanced effects of PDL on
melasma lesions [34].
The copper bromide laser (Dual Yellow; Norseld) is a laser with a concomitant-
output dual-wavelength light source comprising 90 % yellow light at 578 nm target-
ing vascular lesions and 10 % green light at 511 nm which targets pigmentary
lesions. These two light wavelengths can be emitted separately or simultaneously.
Recent studies have suggested the potential effectiveness of targeting the vascular
component of melasma [33, 34, 35, 37].
In a recent pilot study, ten Korean women with mixed or epidermal melasma were
treated with a copper bromide laser emitting both wavelengths simultaneously at
2-week intervals for a total of 8 weeks. MASI scores decreased modestly from an
average of 12.3 pretreatment to 9.5 at 1-month posttreatment follow-up. Using a
chromometer, the authors noted measurable lightening of lesional skin after treatment,
but the effects appeared to wane slightly at 1-month posttreatment follow-up. The
same findings were seen when erythema was measured. Clinically, three patients
were noted to have recurrence at the 6-month posttreatment follow-up. The histo-
logic examination of lesional skin before and 3 months after treatment showed
decreased levels of basal layer melanin (Melan-A) and fewer melanosomes in the
epidermis after treatment, suggesting some longer-term benefit. In addition, CD34
staining for blood vessels showed a decrease in the number and size of dermal ves-
sels after treatment. Staining for endothelin 1 and VEGF antigen showed decreased
numbers in keratinocytes posttreatment, indicating some effect of the laser on vascu-
larity within treated lesions. It is significant to note that none of the ten patients
exhibited scarring or dyspigmentation from treatment. The authors concluded that
the copper bromide lasers seem relatively safe and at least moderately effective for
melasma in Asian patients (level of evidence, II-iii). Additional studies in other
patient populations will help determine the generalizability of these results [35].
A study done in Thailand, however, did not corroborate these positive results.
Among 20 melasma patients treated with the copper bromide laser, the mean mela-
nin index (MI) showed no statistically significant improvement compared with
baseline measurements at any of the follow-up visits. Though there were significant
improvements in clinical evaluation after three treatments (p = 0.00), this difference
was no longer visible after six treatments. There was no improvement as measured
by clinical evaluation or MI. The authors concluded that copper bromide laser does
not improve melasma in patients with skin phototypes III–V [36].
Ghorbel et al. from France conducted a randomized split-face study comparing cop-
per bromide lasers with the triple cream (a combination of hydroquinone, 5 %;
154 C.L. Goh
dexamethasone acetate, 0.1 %; and retinoic acid, 0.1 %). All patients applied the topical
cream to their entire face once a day for 4 weeks. A hemiface was then randomly
assigned to be treated with the copper bromide laser, while the other side of the face
continued to receive daily application of the topical cream for 3 additional months.
Four sessions of copper bromide laser were given at weeks 4, 6, 9, and 12. The yellow
and green wavelengths were simultaneously produced at a ratio of 9:1. The treatments’
effectiveness was assessed using the MASI score for each hemiface. Follow-up visits
were conducted at 3 and 6 months. The main evaluation criterion was the patient’s
MASI score 6 months after the end of treatment. At the end of treatment, the topical
cream resulted in a greater decrease in the MASI score compared with the laser treat-
ment (p = .006). The MASI score at 6 months was comparable with the score at the
beginning of the study in both groups; no significant difference was observed between
the two groups. No difference could be found when results were analyzed according to
the localization and duration of the melasma (p > .99 and p = .87, respectively). An
increased vascularization was noted on the melasma lesions at baseline compared with
perilesional skin. However, no decrease in vascularization was observed on the laser-
treated side between the baseline and posttreatment visits. At the final visit, no changes
in vascularization were noted between the two sides. Neither scarring nor PIH was
noted. The authors concluded that results showed that Kligman’s formula combination
cream is more effective than the copper bromide laser for treating melasma [37].
It would appear that there is little role for the copper bromide laser in the treat-
ment of melasma since the latter two reports appear to indicate lack of superiority
over topical treatment. The role of other vascular lasers remains to be ascertained.
16.3 L
evel of Evidence for Using Lasers and Light
Procedures for Treating Melasma Using the GRADE
Working Group Recommendation [38–40]
16.4 Conclusion
Topical treatment together with sun avoidance and protection remains to be the first-
line treatment for melasma. Lasers and light devices are reserved as second- or
third-line treatments, as they are not curative, do not clear melasma completely, and
may be associated with post-inflammatory hyperpigmentation and relapses. They
do have a role to play in recalcitrant melasma unresponsive to topical treatment,
providing a respite and improvement of quality of life.
IPL appears to be the safest among the laser/light devices. However, it modestly
offers only mild to moderate improvement, requires multiple sessions, and neces-
sitates the need of maintenance treatment. Laser toning is the other alternative, but
complications including rebound pigmentation and guttate hypomelanosis are seri-
ous impediments to watch out for. The QS pigment lasers using standard protocols,
fractional lasers, and copper bromide lasers do not appear to be useful for melasma.
The role of pulsed dye lasers remains unclear. More studies are needed to ascertain
their effectiveness for melasma.
Given their cost and the need for multiple treatments, laser and light therapies
should be considered third-line treatments in severe refractory melasma patients
who have not responded to topical preparations or chemical peels and who are will-
ing to accept the risks of these procedures.
References
1. Na JI, Choi SY, Yang SH, Choi HR, Kang HY, Park KC. Effect of tranexamic acid on melasma:
a clinical trial with histological evaluation. J Eur Acad Dermatol Venereol. 2013;27(8):
1035–9.
2. Tse Y, Levine VJ, McClain SA, Ashinoff R. The removal of cutaneous pigmented lesions with
the Q-switched ruby laser and the Q-switched neodymium: yttrium-aluminum-garnet laser. A
comparative study. J Dermatol Surg Oncol. 1994;20:795–800.
3. Alster TS, Lupton JR. Laser therapy for cutaneous hyperpigmentation and pigmented lesions.
Dermatol Ther. 2001;14:46–54.
4. Taylor CR, Anderson RR. Ineffective treatment of refractory melasma and post-inflammatory
hyperpigmentation by QS ruby laser. J Dermatol Surg Oncol. 1994;20(9):592–7.
5. Manaloto RMP, Alster T. Erbium: YAG laser resurfacing for refractory melasma. Dermatol
Surg. 1999;25:121–3.
156 C.L. Goh
26. Wang CC, Hui CY, Sue YM, Wong WR, Hong HS. Intense pulsed light for the treatment of
refractory melasma in Asian persons. Dermatol Surg. 2004;30:1196–200.
27. Li YH, Chen ZS, Wei HS, Wu Y, Liu M, Xu YY, et al. Efficiency of and safety of Intense pulsed
light in treatment of melasma in Chinese patients. Dermatol Surg. 2008;34:1–9.
28. Bae MI, Park JM, Jeong KH, Lee MH, Shin MK. Effectiveness of low-fluence and short-pulse
intense pulsed light in the treatment of melasma: a randomized study. J Cosmet Laser Ther.
2015;17(6):292–5.
29. Jun WJ, Lee SM, Han JS, Lee SH, Chang SY, Haw S, et al. A prospective, split-face, random-
ized study of the efficacy and safety of a novel fractionated intense pulsed light treatment for
melasma in Asians. J Cosmet Laser Ther. 2015;17(5):259–66.
30. Vachiramon V, Sirithanabadeekul P, Sahawatwong S. Low-fluence Q-switched Nd:YAG 1064-
nm laser and intense pulsed light for the treatment of melasma. J Eur Acad Dermatol Venereol.
2015;29(7):1339–46.
31. Kim EH, Kim YC, Lee ES, Kang HY. The vascular characteristics of melasma. J Derm Sci.
2007;46:111–6.
32. Passeron T. Melasma pathogenesis and influencing factors: an overview of the latest research.
J Eur Acad Dermatol Venereol. 2013;27(S1):5–6.
33. Passeron T, Fontas E, Kang HY, Bahadoran P, Lacour JP, Ortonne JP. Melasma treatment with
pulsed-dye laser and triple combination cream: a prospective, randomized, single-blind, split-
face study. Arch Dermatol. 2011;147(9):1106–8.
34. Passeron T. Long-lasting effect of vascular targeted therapy of melasma. J Am Acad Dermatol.
2013;69(3):e141–2.
35. Lee HI, Lim YY, Kim BJ, Kim MN, Min HJ, Hwang JH, Song KY. Clinicopathologic efficacy
of copper bromide plus/yellow laser (578 nm with 511) for treatment of melasma in Asian
patients. Dermatol Surg. 2010;36:885–93.
36. Eimpunth S, Wanitphadeedecha R, Triwongwaranat D, Varothai S, Manuskiatti W. Therapeutic
outcome of melasma treatment by dual-wavelength (511 and 578 nm) laser in patients with
skin phototypes III-V. Clin Exp Dermatol. 2014;39(3):292–7.
37. Ghorbel HH, Boukari F, Fontas E, Montaudié H, Bahadoran P, Lacour JP, Passeron T. Copper
bromide laser vs triple-combination cream for the treatment of melasma: a randomized clinical
trial. JAMA Dermatol. 2015;151(7):791–2.
38. GRADE Working Group. Grading quality of evidence and strength of recommendations. BMJ.
2004;328:1490.
39. Guyatt GH, Oxman AD, Vist GE, Kunz R, Falck-Ytter Y, Alonso-Coello P, Schünemann HJ,
Grade Working Group. GRADE: an emerging consensus on rating quality of evidence and
strength of recommendations. BMJ. 2008;336:924–6.
40. GRADE Working Group. Rating quality of evidence and strength of recommendations: What
is “quality of evidence” and why is it important to clinicians? BMJ. 2008;336(7651):995–8.