Conflict of Interest Disclosure: All authors have completed and submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest and have disclosed the following: M. Milanic was recipient of a USAF student travel grant to present at ASLMS 2012. The laser equipment was provided by Fotona d.d. for study purposes. Fotona d.d. provided no financial support for these studies, nor did they participate in the design, data collection or analysis of result.
Background and Objectives
Nd:YAP laser emitting at 1,342 nm appears promising for nonablative skin rejuvenation treatment, based on favorable absorption properties of water and melanin in this part of the spectrum. A quantitative determination of energy deposition characteristics of Nd:YAP in normal human skin should enable design of a safe and effective treatment protocol for future human studies.
Energy deposition profile of a prototype Nd:YAP laser was determined using pulsed photothermal radiometry. This technique involves time-resolved measurement of mid-infrared emission from a sample after pulsed laser irradiation. The laser-induced temperature depth profile is reconstructed from the radiometric transients using a custom optimization algorithm, developed and tested earlier in our group. Measurements were performed on the extremities of four healthy volunteers at low radiant exposure (2.8 J/cm2). For the purpose of comparison, energy deposition characteristics of commercial Nd:YAG and KTP lasers (at 1,064 and 532 nm, respectively), were also determined at the same test sites.
On average, the Nd:YAP laser deposits 50% of the absorbed energy within the top 0.36 mm of skin and 90% within 0.86 mm, which is significantly shallower than the Nd:YAG laser. The ratio between the dermal versus epidermal heating is more favorable and shows a smaller inter- and intra-patient variance as compared to both Nd:YAG and KTP laser.
Nonablative photorejuvenation (NAPR) is noninvasive treatment of photoaged skin with intense pulses of light. A variety of lights sources are being used or considered for NAPR, but most common are pulsed lasers emitting in visible or near-infrared (IR) part of the spectrum.
The main mechanisms of NAPR are controlled thermal injury to collagen in papillary and upper reticular dermis and thermal activation of fibroblasts, leading to synthesis of new extracellular matrix. It is therefore believed that light-induced heating effects should ideally be confined to subsurface depths of 0.1–0.5 mm 1, while sparing the epidermis and subcutaneous tissues from nonselective thermal damage.
In most investigational protocols, the depth of thermal injury is assessed by histology 2, 3. However, such assessment of injury depth is very labor intensive and somewhat subjective, as well as prone to artifacts originating from preparation of histological samples (shrinking and deformation) and variation of skin structure between different biopsy sites. Moreover, because this procedure is invasive, researchers often resort to less reliable animal models.
In present study we rely on the fact that, for the purposes of NAPR, the most relevant characteristic of light-tissue interaction is the induced temperature depth profile. From the latter, local thermal damage will develop (or not) according to universal laws of heat diffusion and protein denaturation kinetics.
We determine the laser-induced temperature depth profiles using a less-known technique, called pulsed photothermal radiometry (PPTR) 4, 5. PPTR is based on time-resolved measurement of mid-infrared (IR) emission after pulsed irradiation of a sample. From such radiometric transients, the initial laser-induced temperature profile is reconstructed using a custom optimization algorithm, developed and tested earlier in our group 6, 7. This approach allows noninvasive comparison of energy deposition characteristics from different lasers in the same test site 8.
Our study focuses on temperature depth profiles induced in healthy human skin by a prototype Nd:YAP laser emitting at 1,342 nm. Based on absorption properties of water 9, the main constituent of skin and primary chromophore in IR part of the spectrum (see Fig. 1) 10–12, this wavelength appears very promising for NAPR treatment. In comparison with the common Nd:YAG laser (λ = 1,064 nm), for example the Nd:YAP is expected to cause enhanced heating of the upper dermis at reduced risk of unwanted thermal damage to subcutis. We test this hypothesis by analyzing temperature depth profiles induced by both lasers in normal human skin in vivo.
In the same manner, we characterize also energy deposition achieved with the 532 nm KTP laser, intended primarily for treatment of vascular malformations. Because there is a lot of past clinical experience and many published reports related to interaction of this laser with human skin, these results can serve, in part, as a verification of our PPTR profiling approach.
PATIENTS AND METHODS
Our study involved four healthy volunteers with Fitzpatrick skin type II (all males, 20–50 years old). PPTR measurements were performed on the inner forearm and shoulder of each volunteer. These anatomical locations were selected because they were easily accessible with our experimental setup. In addition, this introduces some variation in skin structure and composition, which increases credibility of our results. Prior to measurement, each test site was shaved and the superficial layer of dehydrated skin cells removed by tape stripping. The site was cleaned using medical-grade ethanol, rehydrated using physiological solution, and air-dried.
Each measurement site was irradiated first with a single 5 millisecond pulse at 1,064 nm, emitted from a medical-grade commercial Nd:YAG laser (Dualis VP by Fotona, Ljubljana, Slovenia). The spot size diameter was approximately 5 mm (Fig. 2; dashed line), and the peak radiant exposure was H0 = 1.46 J/cm2. This value is much lower than the radiant exposures used in any clinical treatment with millisecond Nd:YAG lasers, and did not cause any observable adverse effects or discomfort to our volunteers.
During and following the irradiation, mid-IR emission from the skin in the spectral band of 3.5–5.1 µm was collected by a 50 mm objective onto the InSb array of a fast IR camera (SC7500 by FLIR Systems, Boston, MA). Radiometric images were acquired at a rate of 1,000 frames per second for a total period of 5 seconds. Three measurements were performed at each test site, with an interim interval of at least 1 minute to allow for thermal relaxation of the irradiated skin.
Next, similar measurements were performed on the same skin site using a prototype Nd:YAP laser (Fotona) emitting at 1,342 nm. The pulse length was ∼5 milliseconds, beam diameter just under 5 mm (at half-maximum; see Fig. 2), and central radiant exposure was H0 = 2.83 J/cm2.
In some test sites, we have applied also a commercial 532 nm KTP laser (Dualis VP) with pulse duration of 1 millisecond, hat-top beam profile with diameter of 5 mm, and radiant exposure H0 = 0.38 J/cm2.
IR emission records were analyzed within a 10 × 10 pixel sub-window of the camera array, corresponding to an area of 1.5 mm2 × 1.5 mm2 located near the center of the laser spot. For conversion of radiometric readings to temperature values, we relied on the manufacturer provided calibration setup (Hypercal™) 13. At each pixel, the baseline temperature was subtracted and the remaining values were averaged laterally, to produce a single PPTR signal for each irradiation case.
In the interest of space, the theory of temperature depth profiling using PPTR will not be recapitulated here 4, 5, 14. In a nutshell, PPTR involves measurement of transient changes in mid-IR emission after pulsed irradiation of a sample. From such radiometric signal, we reconstruct the initial laser-induced temperature profile using a dedicated minimization algorithm, developed earlier in our group 6, 7. Detailed numerical simulations and extensive testing in agar and collagen-based tissue phantoms demonstrated that our PPTR setup enables accurate localization of subsurface absorbers 6, 7. Most recently, several in vivo applications confirmed that the described approach provides a unique and valuable insight into specifics of pulsed-light interaction with human skin 8, 15–17.
In terms of specifics, the temperature profiles presented below were reconstructed using non-uniform signal binning 18 and our custom reconstruction code involving a spectrally composite kernel matrix 6 and so-called υ-method minimization algorithm with a non-negativity constraint 6. The assumed tissue parameters (thermal diffusivity D = 0.11 mm2/seconds; reduced heat transfer coefficient h = 0.02 mm−1) and infrared properties were the same as reported before 7, 8. The resulting profiles consist of 100 temperature values, distributed nonuniformly across a total depth of 2 mm.
Figure 3 presents PPTR signals as measured on the shoulder of one volunteer after irradiation with a single pulse from the Nd:YAG, Nd:YAP, and KTP laser, respectively. For the purpose of comparison, all signals are scaled to the same radiant exposure value, H0 = 10 J/cm2.
All signals reach the maximal value immediately after irradiation (time t = 0), and afterwards relax monotonically toward the baseline level (set to 0 by definition). The prominent differences between the initial signal amplitudes reflect primarily the absorption properties of the epidermis at the three involved wavelengths (see Fig. 1). In addition, the different time evolutions at later times bear information on spatial distribution of deposited energy deeper inside the skin.
Figure 4 presents the initial temperature profiles induced by the customary Nd:YAG laser, as reconstructed from PPTR measurements in forearm and shoulder of the same volunteer. Note that the temperature values are scaled up to a hypothetical radiant exposure of 10 J/cm2. Both profiles feature a pronounced temperature peak at subsurface depth of 30–40 µm, which results from absorption of the 1,064 nm light in epidermal melanin and water. The smaller peaks, located at 0.25–0.30 mm (arrow), arise from absorption in dermal water and blood in the upper vascular plexus. The differences between the two profiles are limited to the amplitude of the epidermal temperature peak and presence of secondary “shoulder” (0.60–0.70 mm deep) in the profile from the shoulder test site, which is not seen in the forearm.
Temperature profiles reconstructed from measurements on the same test sites following irradiation with the 1,342 nm Nd:YAP laser are presented in Figure 5. The two profiles obtained from different anatomical locations (see the labels) are very similar to each other. The epidermal peaks are located at the same depth as in the former example (Fig. 4), while the dermal peak has shifted to a somewhat shallower location. It is evident that the temperature profile induced by the Nd:YAP laser decreases more rapidly toward the deeper dermis as compared to the case of Nd:YAG irradiation at 1,064 nm. Also, the epidermal temperature peak relative to the temperature rise in the upper dermis is significantly smaller than with the Nd:YAG laser.
In Figure 6, we present temperature profiles induced in the same test sites by the green KTP laser. As could be expected from the absorption spectra (Fig. 1) and PPTR signal amplitudes (Fig. 2), these profiles feature a much higher epidermal temperature peak. This is followed by a very rapid decrease of the temperature rise into the dermis, resulting primarily from the much higher absorption in blood at 532 nm as compared to the two IR wavelengths. Temperature profile from the volunteer's shoulder (black line) includes a relatively small temperature rise at depths beyond 1 mm. This could be attributed to presence of a subcutaneous blood vessel, which was not detected with the two IR lasers due to the much lower absorption coefficient of blood at the respective wavelengths (Fig. 1).
To enable quantitative analysis of the results, we determine from each reconstructed temperature profile the subsurface depth, up to which 50% of the absorbed laser energy was deposited, z50. The average values and standard deviations, presented in Figure 7 for each combination of anatomic location (shoulder and forearm) and irradiation laser, are computed from three successive measurements performed in all volunteers. For both test locations, the z50 values obtained with the Nd:YAP laser (gray) are evidently smaller in comparison with those for Nd:YAG (white). The values for the KTP laser are smaller still (albeit perhaps not significantly so), but this is of little relevance for the discussed application.
Between the two measurement sites, the results in Figure 7 do not differ beyond their standard deviations (vertical bars) for any laser under test. In Table 1, we, therefore, present the average values of z50 and their standard deviations, as obtained from measurements in all volunteers and both anatomic locations together (i.e., up to 24 temperature profiles for each laser). In our analysis, the 1,342 nm Nd:YAP laser on average deposits 50% of the absorbed energy into the top 0.36 mm of skin, which is 33% shallower as compared to Nd:YAG.
Table 1. Energy Deposition Parameters for the Nd:YAG, Nd:YAP, and KTP Laser
Laser (wavelength) (nm)
αepi (K cm2/J)
αder (K cm2/J)
The presented average values and standard deviations were assessed from all measurements performed with each laser (see the text for definitions).
0.54 ± 0.08
1.22 ± 0.24
0.43 ± 0.07
0.24 ± 0.05
1.8 ± 0.5
0.36 ± 0.06
0.86 ± 0.17
1.75 ± 0.19
1.50 ± 0.16
1.2 ± 0.2
0.27 ± 0.08
0.88 ± 0.23
4.9 ± 2.0
1.7 ± 0.5
2.8 ± 1.4
We have also determined the depths up to which 90% of the absorbed laser pulse was deposited with each laser (z90; second column). The average value obtained for the Nd:YAP laser is 0.86 mm. In comparison, the value for the Nd:YAG laser (z90 = 1.22 mm) is 42% higher.
In addition to the penetration depths (discussed just above) it is also of interest to analyze the temperature rises induced by the three lasers. In Table 1, this characteristic is expressed by two specific heating parameters, evaluated separately for the epidermis (αepi) and dermis (αder). The first represents the peak temperature value in the epidermis (i.e., the shallowest subsurface peak in reconstructed temperature profiles) divided by the applied radiant exposure. Specific heating of the dermis (αder) is determined in a similar way, from the amplitudes of the secondary peaks or “shoulders” beyond the typical epidermal thickness of 0.1 mm.
The results show that both αepi and αder are several times larger for the Nd:YAP laser as compared to Nd:YAG. This is in agreement with considerably stronger absorption coefficient of water at 1,432 nm in comparison with 1,064 nm (Fig. 1).
From the point of view of thermal treatments targeting the dermis, such as NAPR, however, the most important figure of merit is the ratio between the epidermal and dermal heating. As can be seen from Table 1 (last column), the ratio αepi/αder is lowest for the Nd:YAP laser. This suggests that it should be considerably easier to achieve a desired heating of the upper dermis while avoiding epidermal injury with the Nd:YAP, as compared to the Nd:YAG laser which features a 50% higher value of the ratio αepi/αder.
For completeness, Table 1 presents all energy deposition parameters also for the KTP laser. Of course, this does not imply that this laser should be considered as a viable candidate for NARP. On the contrary, the very large ratio, αepi/αder = 2.8, only corroborates the general understanding that this is definitely not the case.
In our results (Fig. 7 and Table 1) the 1,342 nm Nd:YAP laser on average deposits 50% of the absorbed energy within a depth of 0.36 mm, and approximately two thirds of the energy up to 0.5 mm deep. This makes it very suitable for controlled heating of the papillary and reticular dermis, as required for non-ablative photorejuvenation (NAPR) 1.
Moreover, 90% of the absorbed Nd:YAP laser energy is deposited within the top 0.7–1.0 mm of skin. Since human skin thickness is typically in the same range or even thicker 19, almost all the deposited energy is thus contained within the skin. In contrast, 10% of the Nd:YAG radiation penetrates deeper than 1.0–1.5 mm, which raises some concerns with regard to nonselective thermal damage to adipose, muscle and cartilage in the subcutis.
Another important difference between the Nd:YAP and Nd:YAG laser irradiation of skin is more than 6× higher specific heating of the dermis with the former (αder = 1.5 K cm2/J; see Table 1). Just like the reduced penetration depth discussed above, this is a direct consequence of the markedly stronger absorption coefficient of water at 1,432 nm as compared to 1,064 nm (Fig. 1).
At the same time, the specific temperature rise in the epidermis (αepi) obtained with Nd:YAP laser is only 3.8× as large as that observed with the Nd:YAG (Table 1). Combined with the above, this results in the ratio αepi/αder which is 33% lower for the Nd:YAP laser as compared to Nd:YAG (Table 1). Thus, it should be considerably easier to achieve a desired heating effect in the dermis while avoiding nonselective epidermal injury with the former laser as compared to the latter.
We believe that, for the most part, the above result can be explained by consideration of absorption spectra in Figure 1. At 1,432 nm, the measured absorption coefficient of the epidermis 10 is lower than that of water 9, reflecting its hydration level (typically around 70% or lower) and a negligible contribution from other constituents. At 1,064 nm, conversely, the epidermal absorption exceeds that of water. This indicates a significant absorption by other chromophores, most likely melanin, which is known to feature a monotonically decreasing absorption coefficient throughout the visible and near-IR spectral range 20.
In addition, note that the laser-induced temperature rise is not only proportional to the absorption coefficient, but depends also on local fluence. Specifically, since the scattering coefficient in skin is monotonically decreasing through the visible and near-IR spectral range 20, we can expect light fluence in superficial layers to be lower at 1,342 nm relative to 1,064 nm (at a given radiant exposure). This effect helps to decrease the ratio of respective αepi values (Table 1) below the ratio of the two-epidermal absorption coefficients (Fig. 1).
It is also interesting to note that the Nd:YAP laser features not only the lowest ratio αepi/αder but also the smallest standard deviation of the same (Table 1), while both values are highest for the KTP. This agrees with the above notion that the 1,342 nm light is least affected by inter- and intra-patient variations in epidermal melanin content. When applied to NAPR, this laser might thus require less adaptation of treatment parameters on individual patient basis, as compared to Nd:YAG and especially the KTP laser.
One important implication of the above relations is illustrated in Figure 8, which compares temperature profiles measured in the same test site following irradiation with the Nd:YAP versus the Nd:YAG laser. The two profiles were scaled independently to reach the maximal epidermal temperature rise of 40°C, which roughly corresponds to the temperatures required to induce epidermal thermal damage 21, 22. This gives us some idea about the temperature profiles which could be induced in that test site without causing adverse effects related to epidermal thermal injury (e.g., dyspigmentation, scarring).
Within this framework, we see that the temperature rise achievable with the Nd:YAP laser (solid line) is 50–100% higher throughout the upper dermis, as compared to that induced by Nd:YAG (dashed). Even so, the latter causes significantly stronger heating beyond the depths of ∼0.8 mm and further into subcutis. Such deep heating appears unnecessary from the point of view of NAPR, and should probably be avoided.
The legend in Figure 8 indicates the radiant exposure values, to which each of the two profiles was scaled to reach the epidermal temperature rise of 40°C. By using the same criterion and average specific heating values (αepi) from Table 1, the thresholds for epidermal injury with Nd:YAG, Nd:YAP, and KTP laser, respectively, can be estimated as 93 ± 15, 23 ± 3, and 8 ± 3 J/cm2.
These threshold values match the existing clinical experience with Nd:YAG and KTP lasers rather well 22, 23. Note that all subjects in our study had rather fair skin and the measurements were performed in January, when epidermal melanin content would have been at the seasonal minimum. Consequently, the measured temperature values were likely near the lower end of the range expected for general population, and the resulting damage threshold estimates might be near the upper end of their respective range. Again, such variations are expected to be largest for the KTP laser, because melanin concentration will affect epidermal absorption at 532 nm much more than at the two IR wavelengths.
Clearly, the presented rudimentary approach can yield only rough estimates of epidermal damage thresholds. For example, performing NAPR with laser pulses significantly longer than 5 milliseconds will allow for partial thermal relaxation of the epidermis during irradiation, thereby enabling safe application of higher radiant exposures. Moreover, most cutaneous laser applications nowadays involve active precooling of skin surface 24, 25. This can increase epidermal damage thresholds for both the Nd:YAG and KTP laser far beyond the values estimated above, depending on selected technology and implementation details.
The same considerations will certainly apply also for the Nd:YAP laser and should be taken into account when planning human trials with such a device. Moreover, when the selected active cooling technology provides sufficient protection against epidermal injury, the obtained value of αder (Table 1) can be used to estimate the radiant exposure that should lead to the desired temperature rise in upper dermis.
More accurate predictions can be obtained from numerical modeling of heat diffusion dynamics and protein denaturation kinetics in the involved tissues. By using laser-induced temperature profiles as initial conditions, such a model can predict thermal damage distribution inside the skin at any radiant exposure value 15. In a recent study, we found a very good correlation between the predicted damage scores, based on diagnostic PPTR measurements, and severity of thermal injuries observed in the same test spots in human volunteers 25.
For Nd:YAG laser irradiation, our results indicate that essentially all the absorbed energy was deposited within the top 2 mm of skin (Fig. 4). This appears to be in stark contrast with coagulation depths of 5–6 mm, determined earlier by histology 2, 3. However, those studies involved much longer irradiation times, up to 1 second, and even multiple pulses, both of which can increase the depth of thermal damage by a substantial factor. In addition, they were performed on guinea pig ears ex vivo, the structure and optical properties of which may differ from that of human skin in vivo.
In the end, it is fair to discuss also some inherent limitations of our PPTR profiling approach. This technique is known to be less sensitive to temperature rises deeper within the tissue 5, 6. In quantitative terms, the amplitude of PPTR signal components originating from discrete subsurface absorbers decrease inversely proportionally with their depth, and may eventually get buried in experimental noise.
In addition, it takes time t = x2/(2D) for the heat wave originating from an absorber at depth x to reach the maximal amplitude at the sample surface, where it is picked up by our radiometric setup. For skin with heat diffusivity D = 1.1 × 10−7 m2/second 26, this means that PPTR signal components originating from layers deeper than ∼1 mm will not reach their respective maximal value within our acquisition time of 5 seconds. This further reduces their chance to be distinguished from noise and causes temperature values at deeper locations to be underestimated in the reconstructed profiles.
In our study, only Nd:YAG laser penetrates significantly deeper than 1 mm into skin. The deeper parts of related profiles (Fig. 4) may thus suffer from the discussed artifact, leading to underestimation of its energy deposition depth. Meanwhile, temperature profiles up to ∼1 mm deep should not be significantly affected, so our analysis of Nd:YAP irradiation of skin and its suitability for NAPR remains valid.
Temperature depth profiling based on PPTR measurements provides a practical means for objective and quantitative assessment of laser-induced heat deposition in vivo. Our results indicate that Nd:YAP laser emitting at 1,342 nm is very suitable for applications involving controlled heating of the upper dermis, such as nonablative regeneration of photo-aged skin (NAPR). The risks of adverse effects due to overheating of epidermis or subcutis are significantly reduced in comparison with the 1,064 nm Nd:YAG laser. For persons with fair skin, who can tolerate radiant exposures around 95 J/cm2 at 1,064 nm, or 8 J/cm2 at 532 nm (at 5 milliseconds pulse duration), epidermal damage threshold for Nd:YAP irradiation is expected to be around 20 J/cm2. However, application of active cooling and somewhat higher dosages may be required for safe and effective NAPR.
The authors thank Fotona d.d. for the loan of laser equipment, and the volunteers (L.V. and T.M.) for their kind participation in the study.