Evaluating Real-world Use and Adverse Events From 3262 Patients Treated With 18,203 Cycles of Cryolipolysis for Localized Fat Reduction: A Multilocation Practice Retrospective Chart Review

Abstract

Background

Cryolipolysis is an established method for noninvasive focal subcutaneous fat reduction.

Objectives

The aim of this study was to highlight the safety and real-world use of this technology.

Methods

The electronic medical records of all patients treated with a commercially available cryolipolysis technology (CoolSculpting and CoolSculpting Elite, Allergan Aesthetics, Irvine, CA) between January 2016 and June 2023 at a multilocation group practice were retrospectively reviewed. Extracted procedure-related data included number of total sessions and cycles, number of sessions and cycles per body area, and treatment-related adverse events. A cycle was defined as 1 applicator treating 1 body area.

Results

In total, 3262 patients (2797 female, 465 male) treated with 18,203 cycles across 6245 sessions were included. The mean [standard deviation] age at first treatment was 45.0 [12.8] years (range, 15-83 years). The mean number of cycles per patient was 5.6 [6.5] (range, 1-177), with a median of 4 cycles per patient. The mean number of sessions per patient was 1.9 [1.7] (range, 1-38), with a median of 1 session per patient. The most commonly treated area was the lower abdomen (n = 1761, 4734 cycles), and the most frequently treated body area combination was the upper/mid + lower abdomen (n = 937, 5140 cycles). Dual submental treatment was more common than single applicator therapy. Eighty-seven patients exhibited 180 adverse events (2.05% of cycles), including 3 body areas (n = 2) with paradoxical adipose hyperplasia.

Conclusions

This retrospective chart review of cryolipolysis treatment, among the largest to date, demonstrates its overall safety and broad use across numerous body areas. The paradoxical adipose hyperplasia risk per cycle is between 0.018% (1 in 5501) and 0.048% (1 in 2063), depending on the calculation method.

Level of Evidence: 4 (Therapeutic)

Cryolipolysis—or its vastly more appropriate moniker, cryoadipolysis¹—is one of multiple extant treatment modalities for noninvasive reduction of focal adipose deposits. Given that triglycerides irreversibly crystallize within subcutaneous adipocytes at approximately 10°C, a temperature substantially greater than the freezing point of tissue water (0°C), subcutaneous adipose tissue can be preferentially targeted with controlled thermal diffusion, sparing adjacent water-rich cutaneous and muscular layers.² Although the initial animal model study of this technology demonstrated cold-induced lipid crystallization with no gross evidence of adipocyte damage immediately postexposure, suggesting adipocyte apoptosis as the primary mechanism of cellular damage, the etiologic interplay of apoptosis and necrosis has been proposed by in vitro research.^3,4 Two subsequent ex vivo studies,^5,6 however, have only illustrated strictly apoptotic sequelae, with one showing fragmented, anisometric, and dysmorphic adipocytes with macrophage infiltration and collagen fusion, fragmentation, and hyalinization (including interlobular fibrous septa dissolution) at Days 15 and 45, with attenuated but not completely normalized histologic features at Day 60.⁵ Fibrous septal thickening and inflammatory resolution predominate during Days 60 to 90.⁷

Cryolipolysis is currently performed using either vacuum applicators, which suction tissue into cooled cups (or between cooled parallel plates), or a nonvacuum-conformable-surface applicator for anatomic areas not amenable to suctioning, such as the outer thighs (OTs). Studies have shown that cryolipolysis is safe and not associated with alterations in serum lipid levels or peripheral nerve function.^8,9 A systematic review found erythema to be the most common posttreatment adverse event (AE), followed by transient numbness/paresthesia, bruising, and edema.¹⁰ Delayed or persistent pain and peripheral nerve-related AEs are rare. Paradoxical adipose hyperplasia (PAH), a delayed marked increase in adipose tissue localized to the site of applicator placement, has also been rarely reported.¹¹

We sought to highlight the safety and real-world use data of this technology in this retrospective chart review.

METHODS

Patient and Data Procurement

We performed a retrospective chart review of all patients treated with a commercially available cryolipolysis technology (CoolSculpting and CoolSculpting Elite, Allergan Aesthetics, Irvine, CA) between January 1, 2016 and August 1, 2023 at a multilocation group practice. All patient chart data were extracted from a cloud-based electronic medical record system.

Demographic data included age and gender. Race or ethnicity were not consistently reported and were therefore not evaluated. Procedure-related data were also extracted and included the number of total sessions, number of total cycles, number of sessions and cycles per body area, and any AE information during or following treatment. The number of treatment cycles for each patient was validated by comparing medical records with billing data. AEs (both provider- and subject-reported) were diagnosed or confirmed by either dermatologists, nurse practitioners, or physician assistants.

Treatments

Treatments were performed at 11 physical locations of a single practice by a combination of dermatologists, nurse practitioners, and physician assistants, all of whom had completed the relevant device training and were certified by the device manufacturer. Users performed all treatments according to the on-label user guidelines, manufacturer instructions, and clinical best practices. A cycle was defined as 1 applicator to 1 body area, with patients having the option to undergo multiple cycles during the same visit in the same (overlapping) area or different body areas.

A combination of first-generation (CoolSculpting, 2016-2017), second-generation (CoolSculpting, 2017-2021), and third-generation (CoolSculpting Elite, 2021-2023) applicators were utilized for treatment. The first generation consisted of CoolCurve+, CoolCore, CoolMax, CoolFit, and CoolSmooth applicators. The second generation included CoolAdvantage, CoolAdvantage Plus, CoolAdvantage Petite, CoolMini, and CoolSmooth Pro applicators. The third generation included curve (4 sizes), flat (2 sizes), and smooth series applicators.

Written informed consent was obtained from all patients, and all questions were answered prior to treatment. Each treatment area was marked in a standing position and prepped with alcohol pads. With the patient in a recumbent position, a gel pad was placed on the treatment area followed by the appropriate applicator. Upon cycle completion, the applicator was removed, with a successful full draw confirmed by a combination of gel trap imprint, erythema, and cool skin within the marked area. A 2-minute vigorous massage was performed immediately following applicator removal.

Data Analysis

Descriptive analyses of patient charts were performed. Cycles were only recorded if fully completed; incomplete cycles repeated due to device error or patient issue were counted as 1 cycle. Reports of patient satisfaction or dissatisfaction were translated into 1 (satisfied) or 0 (unsatisfied) scores. AE rates were reported based on the total number of treatment cycles with ≥3-month follow-up. Mean, standard deviation, range, and median were computed for quantitative variables, and frequency (percentage) was computed for nominal variables. Standard error of skewness was utilized to test normality for mean comparison. Mean session and cycle numbers between groups and mean treatment interval data within groups were compared using a Mann-Whitney test for nonparametric data, as was satisfaction between groups. The level of statistical significance was set at P < .05.

RESULTS

A total of 3262 patients (2797 female, 465 male) treated with a total of 18,203 treatment cycles and 6245 treatment sessions were included in this retrospective chart review. The mean [standard deviation] age at first treatment session was 45.0 [12.8] years (range, 15-83 years), with a median of 44 years. The mean number of treatment cycles per patient was 5.6 [6.5] (1-177), with a median of 4 cycles per patient. The mean number of treatment sessions per patient was 1.9 [1.7] (1-38), with a median of 1 treatment session per patient. A single body area was treated in 1372 patients, while 1889 patients received treatment of 2 or more body areas. Patient, session, and cycle numbers separated by body area are detailed in Table 1, as are mean and median interval times between treatment sessions. Mean maximum follow-up duration was 11.4 [17.4] months (range, 0-90.7 months), with a median maximum follow-up of 5 months.

The mean number of sessions per patient was significantly greater for the submentum than the upper/mid abdomen (1.5 [0.7] vs 1.4 [0.8], P = .0076) and significantly lower for the upper/mid abdomen than the lower abdomen (1.4 [0.8] vs 1.5 [0.8], P = .0050), waist/flanks (1.5 [1.0], P = .021), and bra-back (1.5 [1.2], P = .035). No further significant differences in mean number of sessions per patient were demonstrated. The knees showed a trend towards the largest number of sessions per patient (2.0 [1.6]) but this was not statistically significant. The mean number of total cycles per patient was significantly lower for the submentum (2.3 [1.3]) than multiple body areas, including the upper/mid abdomen (2.7 [1.8], P < .0001), lower abdomen (2.7 [1.6], P < .0001), waist/flanks (3.2 [2.1], P < .0001), bra-back (3.3 [2.6], P < .0001), arms (2.9 [1.6], P < .0001), inner thighs (3.0 [2.5], P < .0001), and outer thighs (2.7 [2.0], P = .0018). The mean number of total cycles per patient was also significantly lower for the upper/mid and lower abdomen than the waist/flanks (both P < .0001), bra-back (both P < .0001), arms (P = .022 and P = .013, respectively), and inner thighs (P = .021 and P = .016, respectively), but significantly greater for the waist/flanks and bra-back than the arms (P = .0037 and P = .012, respectively) and outer thighs (P = .0003 and P = .0009, respectively)

Although the lower abdomen was the most commonly treated area (n = 1761, 4734 cycles), the submental area was most often treated exclusively, with 455 patients undergoing cryolipolysis only of this region (1063 cycles). Other areas treated exclusively are listed in Table 2. The most frequent combination of body areas treated in individual patients was the upper/mid + lower abdomen (n = 937, 5140 cycles), with other areas treated in combination delineated in Table 3.

Submental treatments were performed with 1 to 3 applicators per session. Dual or triple therapy involved 2 or 3 overlapping cycles, respectively, sequentially during the same session (albeit at times across separate sessions). Dual submental treatment was the most common (n = 349 with 466 sessions), followed by single applicator treatment (n = 312 with 431 sessions), and with triple therapy being relatively uncommon (n = 8 with 11 sessions).

Multiple body regions had a significantly lower mean time interval between the first and second sessions than between the second and third treatment sessions. These included the submentum (123.6 [229.9] [median, 42] vs 264.8 [356.8] [median, 112] days, P = .012), upper/mid abdomen (111.3 [199.3] [median, 46] vs 264.0 [405.9] [median, 127] days, P = .012), lower abdomen (115.0 [218.1] [median, 42] vs 269.1 [348.0] [median, 121] days, P < .0001), and waist/flanks (126.2 [277.5] [median 42] vs 229.9 [321.3] [median 114] days, P = .0047). A significantly lower time to retreatment was seen between the first and second sessions than between the third and fourth treatment sessions in the lower abdomen (115.0 [218.1] [median, 42] vs 200.7 [248.4] [median, 94.5] days, P = .015). No other significant treatment interval differences were demonstrated. The average median interval between sessions for all areas was 52.0 [7.2] days (43.4 [5.3] days between the first and second session, 111.4 [19.9] days between the second and third session, and 100.3 [20.6] days between the third and fourth session).

Satisfaction

Data regarding treatment satisfaction were reported for only 647 individual body areas (10.0%). Nevertheless, 86.4% (559/647) of these patients were satisfied, with a mean score of 0.86 [0.34]. Site-specific satisfaction data are listed in Table 1. There were no statistically significant differences in satisfaction scores between body areas.

Adverse Events

A minimum of 3-month follow-up data were available for 2785 patients with a total of 16,504 cycles (3465, 9162, and 3877 cycles for first-, second-, and third-generation devices, respectively). AE frequency and duration (or time to diagnosis) are detailed in Table 4 and further subcategorized by body area in Table 5. These include a total of 87 patients (3.12%) with 180 AEs, although 16 patients account for 43.3% (78/180) of these—including 1 unfortunate patient with 16 AEs—and 338 implicated cycles (2.05%).

Procedural pain and anxiety/vasovagal symptoms (particularly nausea and lightheadedness) were infrequent and reported to be significant in 14 patients/47 cycles and 14 patients/24 cycles, respectively, which are further separated into treatment-related and massage-related in Tables 4 and 5. No treatments were discontinued due to procedural pain. Two patients were given periprocedural ibuprofen, and 1 patient was prescribed a short course of postprocedure gabapentin 300 mg po TID. Only 2 patients (both submental) required treatment discontinuation as a result of anxiety/vasovagal symptoms.

Treatment for acute posttreatment pain and tenderness was recommended in 50% (8/16) of patients and included tylenol or ibuprofen (62.5%, 5/8), gabapentin (25.0%, 2/8), or a combination of gabapentin and tramadol (12.5%, 1/8). Four patients experienced delayed neuropathic pain beginning at least 48 hours posttreatment, of whom 3 were treated with gabapentin (300-900 mg po qHS to tid) for a mean of 1.5 [0.9] months (range, 0.5-2 months).

Sixty-one thermal events (0.33%) were recorded, of which 9 were reapplied and successfully completed in the same session, 41 were rescheduled and subsequently successfully completed, 8 occurred near the end of the cycle and were not rescheduled, and 3 (all posterior arms) were unable to be completed and not rescheduled. The most common locations for thermal events were the submental area (29.5%), posterior arms (21.3%), waist/flanks (16.4%), lower abdomen (11.5%), and OTs (8.2%). Seventeen of these thermal events were associated with a subsequent documented AE. Thermal events were associated with 60.0% (6/10) of prolonged erythema, 25.0% (1/4) of cutaneous cryoinjury, 25.0% (2/8) of prolonged edema, and 50.0% (4/8) of postinflammatory hyperpigmentation cases.

Prolonged erythema and edema were self-limiting, with treatment recommended for only 40.0% (4/10) and 12.5% (1/8) of patients, respectively. For those patients, medium-potency topical corticosteroids were prescribed for erythema and heat/massage was advised for edema. No patient with prolonged localized numbness required treatment. Previous response was not necessarily predictive of repeat outcome, in that 2 patients with prolonged numbness after 1 submental cycle did not experience numbness with additional treatment, and another patient only experienced numbness after their second submental cycle.

Four patients presented with epidermal cryoinjury, 2 with blistering and 2 with superficial erosions. The former were treated with topical corticosteroids (triamcinolone or clocortolone cream), whereas the latter required only occlusive (silicone or petrolatum-based) emollients. No cases of partial or full-thickness necrosis occurred. Postinflammatory hyperpigmentation was treated in 75% (6/8) of patients with either hydroquinone 4% to 6% (37.5%, 3/8), hydroquinone compounded with kojic acid and either tretinoin (12.5%, 1/8) or fluocinolone (12.5%, 1/8), or a nonhydroquinone pigment-correcting topical (25.0%, 2/8).

Cold panniculitis occurred in 9 patients, with 3 patients recommended heat/massage to aid in resolution and none requiring intralesional or oral corticosteroids. An additional 9 patients reported contour irregularity posttreatment, with the majority (77.8%, 7/9) noted to be slight to mild. One case was associated with posttreatment weight gain and 2 cases were related to pre-existing laxity mimicking contour deformity. No treatment was recommended in 4 patients, while intralesional water or deoxycholate injections were recommended in 2 cases of left midabdominal and OT irregularity, but this treatment was not pursued by patients. One other patient with mild bilateral inner thigh indentation after 2 sessions (4 cycles) responded well to 1 additional round of cryolipolysis. A moderate/significant contour deformity was seen in 2 patients (22.2%), both involving the OTs. One patient noted worsening bilateral irregularity after 3 sessions (6 cycles), which failed to improve with intralesional water injections. The second patient demonstrated left OT atrophy in the shape of the applicator plate with a visible drop-off after 1 session (1 cycle), requiring tumescent liposuction of the surrounding previously untreated tissue.

PAH was diagnosed in 2 patients based on physical exam and clinical progression, although neither were biopsy confirmed. Both were male, Fitzpatrick Skin Type II, and of European ethnic origin. One was a 49-year-old who presented with marked focal subcutaneous enlargement over a period of 8 months following a single session for the flanks (2 cycles) and lateral chest/upper lateral back (2 cycles) areas using third-generation applicators. Tumescent liposuction was performed, but the areas regained nearly 75% of their abnormal size within 5 months. A second session of tumescent liposuction was performed, with no recurrence noted at 4-month follow-up. The second patient was a 62-year-old whose lower abdomen was treated with 2 sessions (4 cycles) using first-generation applicators, leading to rapid focal enlargement of the area within 2 months. His waist/flanks had been treated concurrently, but developed no abnormalities. Tumescent liposuction was recommended but not pursued by the patient.

Several other unanticipated AEs were also reported. Irritant contact dermatitis to the gel pad was suspected in 2 patients. One patient with sensory changes following dual submental treatment reported itching and burning that resolved over several weeks with use of petrolatum ointment. Another patient had a localized area of the lower abdomen affected, requiring 1% hydrocortisone cream and oral antihistamines, but which did not recur with subsequent treatments. Painful reactive bilateral axillary lymphadenopathy occurred after a single bra-back and arm treatment (4 cycles total) in 1 patient, but resolved within 2 to 3 weeks. Another patient with a history of deep-vein thrombosis from lower-extremity sclerotherapy developed markedly tender, linear, indurated veins within the treated left proximal inner thigh 2 weeks posttreatment, consistent with superficial venous thrombophlebitis, that resolved with supportive care. Lastly, 3 patients (mean age, 40.0 [3.6] years; range, 36-44 years) with submental treatment developed unilateral lower lip dysesthesia and partial paralysis consistent with injury to the marginal mandibular branch of the facial nerve (MMBFN). Two of these patients had dual treatment, while 1 had a single application. Signs and symptoms in all 3 patients were evident immediately posttreatment and completely resolved by 3-month follow-up.

DISCUSSION

Cryolipolysis remains a popular noninvasive body contouring procedure with predictable outcomes.¹² Numerous prospective studies have demonstrated its safety and efficacy in various body areas, yet few large-scale retrospective studies evaluating real-world use of this technology have been previously reported. A retrospective study of 91 patients (307 cycles) over 13 months at a single center showed 84% patient satisfaction, and patient-reported AEs were limited to redness and pain/discomfort.¹³ Another retrospective review of 528 patients (2729 cycles) over 36 months reported only 3 cases of mild-to-moderate pain or neuralgia.¹⁴ Dierickx et al likewise reported 518 patients (891 body areas) treated over 31 months, with the only side effects being self-limited ecchymosis, erythema, cutaneous sensitivity, and nodular or diffuse induration.¹⁵ The most significant retrospective study to date has been that of Nishikawa and Aikawa, who reviewed 4122 predominantly Asian patients treated over 6 months at 30 clinics.¹⁶ However, none of these studies performed an in-depth analysis into patterns of retreatment and combination therapy.

The most common area of treatment in our patients—in line with previous retrospective reports—was the midsection, accounting for 69.8% of all treatment cycles. Multiple applicator treatments were often performed in the same body area during the same treatment session, particularly in the submental, upper/mid abdomen, and lower abdomen. Our data highlight the fact that patients who receive treatment with this technology are likely to undergo multiple treatments (mean, 5.6 [6.5] cycles), with the majority (57.8%, 1885/3262) seeking treatment of multiple body areas. Patients were also likely to seek retreatment of the same area, with a mean weighted percentage of 32.4% of patients receiving more than 1 round of treatment to the same body area. Initial retreatment, particularly for the submentum and areas of the midsection, was performed relatively quickly (median, 42-46 days), while subsequent retreatment was significantly delayed (median, 112-127 days). Patient satisfaction in this study (86.4% of 647 patients) is relatively consistent with the satisfaction rates of 73% (n = 91) and 84% (n = 518) from previous retrospective surveys.^13,15

AEs were infrequent and typically mild and transient, with only 3.12% of patients (87/2785) demonstrating at least 1 AE, and a per-cycle rate of 1.98% (326/16,504). The most AEs for any particular body area were seen in the lower abdomen (31.7% of all AEs); however, the largest AE to treated patient ratio was for the submentum (5.9%, 36/613), with the lower abdomen appreciably lower (3.2%, 57/1761). A significant fraction of cases of prolonged erythema (60.0%), cutaneous cryoinjury (25.0%), prolonged edema (25.0%), and postinflammatory hyperpigmentation (50.0%) were associated with thermal events, which may be due to inappropriate temperature fluctuations resulting from poor skin-applicator surface contact due to substandard applicator placement, patient movement during treatment, or inherent tissue fibrosity.

A significant number of contour irregularity cases, including both instances of moderate posttreatment deformity, were associated with the OTs. Muscle bulging due to fascia and soft tissue laxity in this area may simulate a subcutaneous fat bulge, with the fibrous, poorly pinchable nature of fat innate to this area further muddying the waters. An inexperienced provider not adept at recognizing this clinical finding may pursue inappropriate treatment with cryolipolysis, producing undue subcutaneous atrophy of the lateral thigh bulge that is clinically disproportionate to its surroundings.

Atypical delayed neuropathic pain has been previously reported and was rare, mild, and short-lived in our patients.^17,18 Given that cryolipolysis can produce transient ​​decreases in the densities of dermal myelinated and epidermal nerve fibers, leading to self-limited alterations of cutaneous sensory nerve function,^8,19 it is not surprising that numbness, itching, paresthesias, or delayed neuropathic pain would occur in a small cohort of patients. However, only 1 case report of transient MMBFN palsy has been previously reported,²⁰ with no cases of MMBFN injury described in prospective clinical trials of the submentum.^21-27 Unlike the previous report from dual submental therapy in a 60-year-old female, our 3 cases (0.49% of patients or 0.36% of cycles) were younger and without significant lower face laxity, with 1 patient only having a single midline applicator placed. The incidence of transient MMBFN paralysis following submental cryolipolysis may therefore be underreported. Although a case of motor neuropathy from posterior arm cryolipolysis has been previously described, we saw no cases in our patients.²⁸

The most concerning AE due to cryolipolysis, however, remains PAH. Its subcutaneous histopathology demonstrates disorganized adipocytes of varying size/shape that may be combined with perilobular septal fibrous thickening and/or vascular hyperplasia; nevertheless, punch or excisional biopsy may only be necessary in cases where the diagnosis of PAH is not clinically obvious.²⁹ Hyperplasia, hypertrophy, or altered metabolism of pre-existing adipocytes; recruitment of adipocyte progenitor cells (eg, preadipocyte or stem cell populations) triggered by tissue hypoxia; mechanotransduction, wherein growth-promoting signals are incited by mechanical tension, such as in areas of firm, nondistensible, fibrous fat; and adipocyte hyperplasia due to a reduction in local sympathetic innervation have all been purported to play a pathophysiologic role.^29,30 Tumescent liposuction can be an effective treatment option for PAH, as shown in a recent case series of 33 patients (60 body areas) treated with ultrasound-assisted liposuction.³¹ Although no patients were refractory to treatment in that study, rapid clinical recurrence demanding repeat lipoaspiration has previously been reported and was demonstrated in 1 of the patients in this review.³²

The risk per patient was initially thought to be as rare as 1 in 20,000 (0.005%) in 2011, increasing to a manufacturer-reported 1 in 4000 cycles (0.025%) in 2015.³³ However, recent publications have reported higher incidence rates, including a large multicenter retrospective review of 8658 cycles in 2114 patients which reported a PAH risk of 1 in 666 cycles (0.15% or 1/235 patients [0.43%]),¹¹ as well as several other smaller retrospective chart reviews showing 1 in 128 patients (0.78%),³⁴ 1 in 211 patients (0.47%),³⁵ 1 in 149 patients (0.67%),³⁶ 1 in 100 patients,³⁷ and 1/600 cycles (0.17%).³³ Interestingly, the aforementioned large-scale review showed a higher rate of PAH with first-generation applicators (1 in 260 cycles [0.39%] or 1 in 50 patients [2.0%]) compared with second-generation applicators (1 in 2020 cycles [0.05%] or 1 in 882 patients [0.11%]).¹¹

We reported 2 patients with PAH with 3 total body areas and 8 total implicated cycles, for a rate per patient of 0.072% (1 in 1393 patients) and per cycle of 0.048% (1 in 2063 cycles). The risk of PAH with first-generation applicators (1 in 866 cycles [0.12%]) was marginally greater than that of their third-generation counterparts (1 in 969 cycles [0.10%]). Previous studies have reported PAH incidence as the number of PAH-affected body areas divided by the total number of cycles, which would make our rate 1 in 5501 (0.018%) and skew the first- vs third-generation risk significantly (1 in 3465 [0.029%] vs 2 in 3877 [0.052%]); however, we feel reporting this as a factor of the total number of cycles playing a role in PAH development may be a more accurate measure. Both patients were male, conforming with the current consensus that males are at disproportionately higher risk of developing PAH.³⁸ Despite reports that patients of Hispanic ethnic origin may be at greater risk,³⁴ both these patients here were of European descent.

Our retrospective chart review is most likely limited by reporting bias. Treated body areas may thereby have been inadequately recorded in the medical record or have been overly generalized. For example, it is possible that the mons and knees may have been described in general terms as lower abdomen and inner thighs, leading to significant underestimation of cycle counts in those body areas. The number of thermal events may also be far higher than reported due to inconsistent documentation. Moreover, treatment interval data may have been skewed by numerous instances of left and right sides of the body being treated separately on sequential days. In terms of AEs, patients informed prior to treatment about potential common, mild, or transient complications may have not reported their occurrence to the treatment provider, or the providers learning about an anticipated or unanticipated treatment-related AE may have failed to document it. It may also be limited by recall bias, wherein patients subsequently neglected to report significant AEs. Despite a median follow-up of 5 months, follow-up times were widely inconsistent between patients with a left skewed data distribution. It is therefore possible that the development of delayed complications such as PAH could have gone undetected. Lack of standardized criteria for evaluating and grading results and poor reporting of patient satisfaction posttreatment also severely limit the ability of this review to assess overall efficacy.

CONCLUSIONS

We present the largest retrospective chart review of cryolipolysis patients to date based on number of treatment cycles, highlighting the frequency of retreatment and combined body area treatment and demonstrating the safety of this noninvasive technology for targeted reduction of localized areas of subcutaneous adipose tissue. The risk of PAH per cycle is between 0.018% (1 in 5501) and 0.048% (1 in 2063), depending on the method of calculation.

Disclosures

Dr Friedmann is a consultant and investigator for Abbvie, Inc. (North Chicago, IL). The other authors have no competing interests to declare. Abbvie, Inc. or its subsidiary, Zeltiq Aesthetics (Pleasanton, CA), played no role in the design, data gathering, or creation of this manuscript.

Funding

The authors received no financial support for the research, authorship, and publication of this article.

REFERENCES

Author notes