Abstract
Objective
To determine the impact on skin perfusion of ice pack application on healthy skin.
Methods
A region of skin on the ventral midline was clipped from 10 healthy dogs. Skin perfusion was assessed with laser speckle contrast imaging at baseline, immediately after 15 minutes of icing, and 15 minutes after removal of the ice pack. Mean speckle contrast was calculated from the laser speckle contrast imaging images at all time points. Mean speckle contrast is a unitless measurement that is inversely proportional to skin perfusion. Local skin temperature was also assessed.
Results
Skin temperature decreased from 92.2 ± 0.6 °F at baseline to 69.1 ± 1.1 °F immediately after icing and had increased to 87.5 ± 0.5 °F at 15 minutes after icing. Skin perfusion decreased (mean speckle contrast increased) by a mean of 10.0 ± 5.7% from baseline to immediately after icing and by a mean of 14.8 ± 7.5% from baseline to 15 minutes after icing.
Conclusions
Application of ice to the skin for 15 minutes decreases skin perfusion for at least 15 minutes after ice pack removal.
Clinical Relevance
Veterinarians should consider that a decrease in skin perfusion is associated with the application of an ice pack when deciding whether to implement this therapy; however, additional research is warranted to determine whether there is an impact on healing.
Following many surgical interventions, the application of ice to the incision site is often recommended to decrease swelling and pain. Anecdotally, icing is avoided following procedures where skin perfusion is of particular concern such as skin flaps or grafts. Such procedures have relatively high complication rates, typically due to necrosis or dehiscence of some portion of the grafted or flapped skin.1–3
Ice packs are most commonly used due to their low cost, convenience, accessibility, and perceived beneficial results.4 Icing decreases inflammation by reducing proinflammatory mediators and increasing anti-inflammatory mediators.5 The ice also reduces blood flow to the area through vasoconstriction, which decreases hydrostatic pressure in the area preventing edema formation, or swelling.5 Analgesia from ice results from slowing the nerve conduction velocity in the local sensory axons.4,5
The duration of ice application necessary to achieve appropriate therapeutic effectiveness in veterinary patients is not well defined. Human studies6–8 found improved outcomes in soft tissue injury healing rates after 10 to 30 minutes of ice pack use with repeated icing intervals. In canine patients, an icing duration of 10 to 20 minutes significantly reduces the temperature of tissues to a depth up to 1.5 cm.4 Pain scores improved in dogs after four 30-minute cold compression therapy sessions in the first 24 hours after tibial plateau leveling osteotomy when compared to those with no icing.6
Laser speckle contrast imaging (LSCI) is a full-field, noninvasive, and noncontact technique to map speckle contrast as an indicator of perfusion.7 Laser speckle contrast imaging acquires data through a laser projected at the skin. The light is scattered and reflected toward the detector in a random interference pattern creating the speckle pattern image.8 This image is then used for analysis. Within tissue, RBC motion scatters the laser thereby creating the speckle contrast. This speckle contrast, or the degree of blurring due to RBC motion, is indirectly related to perfusion.8 Consequently, LSCI can identify blood flow changes and ultimately perfusion changes. As the speckle contrast pattern is affected by movement, motion must be limited when acquiring images. This includes respiration, panting, and general movement of the dog. Although LSCI has been used extensively in human medicine, studies reporting its use in veterinary medicine are limited. Laser speckle contrast imaging has previously been used to assess the impact of dexmedetomidine on skin perfusion in cats.8 It has also been used to assess perfusion following experimental skin flaps in a porcine model.9
The purpose of this study was to evaluate skin perfusion at normal skin temperatures and after ice packing to determine if cooler skin temperatures have an impact on blood flow to the area. We hypothesized that icing the skin for 15 minutes would decrease perfusion of the skin. This study also evaluated the clinical use of LSCI as a noninvasive method to assess perfusion.
Methods
This study was approved by the Texas A&M University’s IACUC (No. 2024-0010 CA). Ten healthy adult client-owned dogs were enrolled in a controlled prospective pilot study. Inclusion criteria included any healthy dog with a calm demeanor such that the dog would tolerate mild restraint for the duration of imaging and icing. At the discretion of the owners, trazodone at 5 mg/kg, PO, was offered the morning of imaging to facilitate stress-free imaging. On the ventral midline, a 4 X 4-cm region of fur was shaved with a No. 40 clipper blade to mimic standard surgical preparation. Local skin temperature was recorded through a skin temperature probe (YSI Compatible Reusable Temperature Probe; part No. 409B; OEM) taped to the clipped area (CareFusion Compatible Disposable, Adhesive Temperature Probe Covers; part No. 0203-1980-300; OEM). The minimum temperature this device can read is 66 °F. Dogs that may have had skin temperatures below this threshold during icing were recorded as a reading of 66 °F.
The LSCI unit (Dynamic Light Inc) was positioned 20 cm away from the shaved site to take images before icing, immediately after icing for 15 minutes, and 15 minutes after removing the ice pack (Figure 1). The LSCI camera has a field of view of approximately 2 cm and takes 100 frames in a 5-millisecond time frame, with the frames automatically averaged into 10 images. Image capture was repeated 3 times in succession to yield a total of 30 images at each time point (before icing, immediately after icing for 15 minutes, and 15 minutes after removing the ice pack).
A to C—During perfusion measurements, dogs were placed in lateral recumbency on an exam table and lightly restrained. A—The camera unit was positioned approximately 20 cm from the shaved site for image acquisition. B—A temperature probe was secured to the shaved skin, and care was taken to include it within the region of the skin being iced. C—To ice the skin, a standard gel ice pack was wrapped in a 1-mm-thick absorbable pad and held by hand over the shaved region with light pressure to mimic incision site aftercare.
Citation: American Journal of Veterinary Research 2025; 10.2460/ajvr.25.01.0023
During perfusion measurements and icing, the dogs were placed in lateral recumbency on an exam table with minimal restraint with hands against the neck and at the hip during image acquisition. To ice the skin, a standard 6 X 7-inch gel ice pack (Comfort Gel Packs; Accurate Manufacturing Inc) was removed from a −0 °F freezer; wrapped in a 1-mm-thick, 12 X 12.5-inch absorbable pad (WypAll; PowerClean; L40 Extra Absorbent Towels); and held to the ventral midline by hand to mimic incision site aftercare (Figure 1). Light minimal external pressure was used to hold the ice pack in place to maximize contact with the skin. Care was taken to ensure the temperature probe was in the center of the icing field during icing. After 15 minutes, the ice pack was removed for immediate LSCI image capture and 15 minutes after icing image capture of the iced area. The local skin temperature at the time of each image capture was reported.
ImageJ software was used for image analysis. Raw speckle images were concatenated into a single stacked image (Figure 2). The Z-projection was applied to this stacked image to visualize the 3-dimensional image data in a 2-dimensional format for analysis. An area of interest that was approximately 65,000 square pixels was manually selected from the center of the image to calculate the mean speckle contrast (MSC). This was repeated for the images taken with each dog at each time point. These MSC values were used for statistical analysis. Mean speckle contrast is inversely proportional to perfusion; an increase in MSC is associated with a decrease in perfusion.
A to C—Series of images obtained during laser speckle contrast imaging. A—Standard white light image from the camera. B—Raw speckle data. C—Stacked speckle image, created in ImageJ and used for mean speckle contrast analysis.
Citation: American Journal of Veterinary Research 2025; 10.2460/ajvr.25.01.0023
Data analysis
Descriptive statistics were reported for this pilot study. Mean speckle contrast, skin temperature, and percent change in MSC were normally distributed based on a Shapiro-Wilk test, while age and weight were not normally distributed. Mean ± SE or median (range) was reported where appropriate.
Results
All enrolled dogs completed the study. There were 7 neutered males and 3 spayed females. The median age was 7.42 years (2.4 to 13.6 years). Median body weight was 31.0 kg (13.6 to 56.7 kg). Breeds represented in the study included Labrador Retrievers (n = 3) and 1 each of a mixed-breed dog, German Shepherd Dog, Golden Retriever, Great Dane, American Pit Bull Terrier, Australian Cattle Dog, and Catahoula Leopard Dog. During the study, a single dose of oral trazodone was administered to 6/10 enrolled dogs at the owner’s discretion. The oral trazodone was administered the night before and the morning of imaging, approximately 2 hours before image acquisition. Four dogs did not receive trazodone.
The mean local skin temperature decreased from 92.2 ± 0.6 °F at baseline to 69.1 ± 1.1 °F immediately after icing and increased to 87.5 ± 0.5 °F at 15 minutes after icing. Mean speckle contrast was increased immediately after icing (0.27 ± 0.02) compared to baseline (0.25 ± 0.02) and further increased at 15 minutes after icing (0.29 ± 0.02). This equates to a decrease in skin perfusion by 10.0 ± 5.7% immediately after icing compared to baseline and by 14.8 ± 7.5% at 15 minutes after icing compared to baseline. When individual values were considered, a decrease in skin perfusion was noted in 7/10 dogs immediately after icing compared to baseline and 7/10 dogs 15 minutes after icing compared to baseline (Table 1).
Individual percent change in mean speckle contrast (MSC) immediately and 15 minutes after application of an ice pack to healthy skin, compared to baseline.
| Dog | Percent change immediately after icing | Percent change 15 minutes after icing | Weight (kg) | Age (y) | Trazodone administered? |
|---|---|---|---|---|---|
| 1 | +21.5 | +23.2 | 31.3 | 10.3 | No |
| 2 | +20.7 | −5.6 | 13.6 | 11.7 | Yes |
| 3 | −3.8 | +13.3 | 33.0 | 13.6 | Yes |
| 4 | −15.1 | −7.5 | 32.6 | 2.4 | Yes |
| 5 | +5.7 | +2.8 | 24 | 4.1 | Yes |
| 6 | +17.3 | +20.0 | 27 | 9.8 | No |
| 7 | +13.7 | +28.6 | 27.2 | 2.6 | No |
| 8 | +46.2 | +71.8 | 56.7 | 7.2 | No |
| 9 | +4.1 | −6.0 | 33 | 5.6 | Yes |
| 10 | −10.5 | +7.7 | 31.6 | 6.9 | Yes |
| Mean ± SE | +10.0 ± 5.7% | +14.8 ± 7.5% |
Cells with decreased perfusion (increased MSC) are in bold and increased perfusion (decreased MSC) are in italics. Data are supplemented with the respective weight, age, and trazodone administration for the 10 enrolled dogs.
Discussion
A decrease in skin perfusion (increase in MSC) was noted after ice pack application to healthy skin. Mean speckle contrast is a unitless measurement that must be compared to its own baseline within each individual dog. As such, a specific MSC value or magnitude of change that correlates with a clinically significant decrease in skin perfusion is difficult to ascertain. This is highlighted by the variation in baseline MSC among individual dogs, although most dogs shared a similar trend. Further investigation is needed to determine how factors such as dog age, skin thickness, and skin pigmentation impact baseline MSC measurements among individuals, as well as what magnitude of decreased perfusion would equate to delayed healing.
Particularly in traumatized skin, a 10% decrease in skin perfusion after icing as noted in this study may be enough to make a clinical difference. For example, a routine skin incision may be able to withstand this degree of perfusion decrease, but skin that is recovering after an axial pattern flap or free skin graft may suffer delayed healing or failure. Additionally, a further decrease in perfusion was noted 15 minutes after icing compared to immediately after icing. Previous work10 in humans supports continued vasoconstriction even after cryotherapy is discontinued. Larger studies in dogs with surgically or naturally traumatized skin should be undertaken to better understand the clinical impact of icing on skin perfusion, in addition to the duration of time it takes skin to return to baseline perfusion.
The relationship between skin thickness, pigmentation level, and LSCI perfusion assessment warrants further evaluation and likely explains the variation in MSC values noted between dogs in this study. Skin thickness varies with breed, species, body region, age, sex, hydration status, and the individual animal itself.11 Skin thickness in dogs ranges from 0.5 to 5.0 mm with individual variation depending on body location.12 This degree of variation highlights the need to compare a baseline for each dog and may hinder the cage-side direct clinical application of LSCI imaging.
Trazodone was used at the owner discretion in this study and thus was not administered to all dogs. Trazodone is a widely used drug administered to decrease measurable signs of anxiety in canine patients.13,14 Trazodone is a serotonin antagonist and reuptake inhibitor that has an anxiolytic effect by blocking the 5-hydroxytryptamine receptor 2A breceptor.13 Although the sedative effects of trazodone in canine patients are documented, the impact of trazodone on skin perfusion has not yet been explored. This warrants further study and must be considered a limitation of the present study.
Several additional limitations were associated with this study. Larger sample sizes are needed for more detailed analysis before clinical application of the findings here. All dogs enrolled in the study were healthy with normal, intact skin; further research in clinical patients with surgically traumatized skin is warranted. Measurements were only taken after one 15-minute session of icing and were not followed out to the return of baseline perfusion. Clinically, icing is typically performed with repeated sessions. Finally, a limitation of LSCI is its sensitivity to motion. Images were captured at the end of expiration to minimize motion artifacts; however, this may have impacted the accuracy of perfusion measurements.
In conclusion, the application of an ice pack to healthy canine skin resulted in a decrease in skin perfusion. Future work is warranted to assess the clinical significance of this finding in a larger cohort of clinical patients.
Acknowledgments
The authors thank the Texas A&M University Soft Tissue Surgery clinicians and technicians for their support during the completion of this work.
Disclosures
The authors have nothing to disclose. No AI-assisted technologies were used in the composition of this manuscript.
Funding
Funded by the Boehringer Ingelheim Veterinary Scholars Program and the College of Veterinary Medicine & Biomedical Sciences, Texas A&M University.
ORCID
V. Dickerson https://orcid.org/0009-0008-0096-9942
References
- 1.↑
Field EJ, Kelly G, Pleuvry D, Demetriou J, Baines SJ. Indications, outcome and complications with axial pattern skin flaps in dogs and cats: 73 cases. J Small Anim Pract. 2015;56(12):698–706. doi:10.1111/jsap.12400
- 2.
Berry CE, Le T, An N, et al. Pharmacological and cell-based treatments to increase local skin flap viability in animal models. J Transl Med. 2024;22(1):68. doi:10.1186/s12967-024-04882-9
- 3.↑
Briers JD. Laser Doppler, speckle and related techniques for blood perfusion mapping and imaging. Physiol Meas. 2001;22(4):R35–R66. doi:10.1088/0967-3334/22/4/201
- 4.↑
Millard RP, Towle-Millard HA, Rankin DC, Roush JK. Effect of cold compress application on tissue temperature in healthy dogs. Am J Vet Res. 2013;74(3):443–447. doi:10.2460/ajvr.74.3.443
- 5.↑
Kunkle BF, Kothandaraman V, Goodloe JB, et al. Orthopaedic application of cryotherapy: a comprehensive review of the history, basic science, methods, and clinical effectiveness. JBJS Rev. 2021;9(1):e20.00016. doi:10.2106/JBJS.RVW.20.00016
- 6.↑
Drygas KA, McClure SR, Goring RL, Pozzi A, Robertson SA, Wang C. Effect of cold compression therapy on postoperative pain, swelling, range of motion, and lameness after tibial plateau leveling osteotomy in dogs. J Am Vet Med Assoc. 2011;238(10):1284–1291. doi:10.2460/javma.238.10.1284
- 7.↑
Kazmi SMS, Faraji E, Davis MA, Huang YY, Zhang XJ, Dunn AK. Flux or speed? Examining speckle contrast imaging of vascular flows. Biomed Opt Express. 2015;6(7):2588–2608. doi:10.1364/BOE.6.002588
- 8.↑
Hollenbeck DL, Simon BT, Scallan EM, et al. Sedation with dexmedetomidine decreases skin perfusion in cats. Am J Vet Res. 2023;84(12):ajvr.23.06.0137. doi:10.2460/ajvr.23.06.0137
- 9.↑
Nguyen CD, Sheikh R, Dahlstrand U, Lindstedt S, Malmsjö M. Investigation of blood perfusion by laser speckle contrast imaging in stretched and rotated skin flaps in a porcine model. J Plast Reconstr Aesthet Surg. 2018;71(4):611–613. doi:10.1016/j.bjps.2017.08.030
- 10.↑
Khoshnevis S, Craik NK, Diller KR. Cold-induced vasoconstriction may persist long after cooling ends: an evaluation of multiple cryotherapy units. Knee Surg Sports Traumatol Arthrosc. 2015;23(9):2475–2483. doi:10.1007/s00167-014-2911-y
- 11.↑
Scott DW, Miller WH, Griffin CE. Muller & Kirk’s Small Animal Dermatology. 6th ed. Saunders; 2001:1–70.
- 12.↑
Affolter VK, Moore PF. Histologic features of normal canine and feline skin. Clin Dermatol. 1994;12(4):491–497. doi:10.1016/0738-081X(94)90215-1
- 13.↑
Jay AR, Krotscheck U, Parsley E, et al. Pharmacokinetics, bioavailability, and hemodynamic effects of trazodone after intravenous and oral administration of a single dose to dogs. Am J Vet Res. 2013;74(11):450–1456. doi:10.2460/ajvr.74.11.1450
- 14.↑
Kim S-A, Borchardt MR, Lee K, Stelow EA, Bain MJ. Effects of trazodone on behavioral and physiological signs of stress in dogs during veterinary visits: a randomized double-blind placebo-controlled crossover clinical trial. J Am Vet Med Assoc. 2022;260(8):876–883. doi:10.2460/javma.20.10.0547
