Van Nuys, CA 91405 7020 Van Nuys Blvd
Visalia, CA 93291 300 E Mineral King 105
free consultation (626) 585-0800

A Review of Liposuction, Abdominoplasty and Facelift Mortality And Morbidity Risk Literature

Powerful circumstantial evidence implicates general anesthesia as the root of why plastic surgeons have had such significantly worse mortality figures for liposuction than specialties who use virtually no general anesthesia, such as dermatologists43, 69-72. Pulmonary embolism is also unquestionably a significant risk in liposuction, accounting for about one out of every four liposuction associated fatalities in two reports8, 60. Pulmonary emboli are strongly associated with general anesthesia, perhaps providing a link between these two most significant risk factors in liposuction.

All three national specialty societies, Plastics, Cosmetic, and Determatology,73-75 have mentioned a five liter aspirate volume in regards to the safety of the procedures in their liposuction protocols. There is broad consensus that (obviously) there exists some liposuction tissue-trauma limit, above which risks must inevitably escalate. The existing literature however, provides no support for this arbitrary aspirate volume figure. A survey of a half-million liposuction surgeries concluded that there was no relationship between fat volume removed and fatal outcome8. Furthermore, the California Medical Board completed an exhaustive review of liposuction safety in 2002 as directed by the legislature, in order to produce regulations. The original pressure to allow larger cases only in hospitals was rejected on the basis of information showing surgi-centers to be as safe - and perhaps safer - than hospital settings. They elected to not regulate volume of liposuction aspirate removed, finding no empirical data to support specific volume limitations. Their decision was based, in part, on insurance company claim records for liposuction-related lawsuits which contrasted outcomes, showing that between 1985 and 1999, there have been three times more malpractice settlements for hospital-based procedures than for office based liposuction73, 76. Though (as Bruner states) hospitals may host more extensive procedures, there are undoubtedly more liposuction surgeries performed in outpatient settings overall73.

Finally, the latest data implicates multiple procedures as a culprit in liposuction mortality. A follow-up ASAPS study surveying a similar group of plastic surgeons who were found to have mortality rates of 1/5,224 in 2000, in which liposuction risk rates for the period between mid-1998 and mid-2000 were calculated, reported risk rates ten-fold less: 1/47,41570. However, in this period of time, it seems unlikely that the procedure, carried out by a sample of the same board-certified plastic surgeons, became that much safer. Particularly unusual is this last study's post-operative nausea and vomiting (PONV) rate of one percent. Such a low rate is inconsistent anesthesia literature. Dozens of studies show PONV rates of 8 to 55 percent for outpatient procedures77-82. This discrepancy suggests under-reporting bias. Regardless, the Young et al report83 went further to calculate that liposuction mortality risk rates increase to 1/7,314 if liposuction is combined with other plastic surgery procedures (a very common event) and rockets to 1/3,218 if combined with abdominoplasty. The latter incidence is 14-fold higher than for liposuction alone and is consistent with the other abdominoplasty mortality literature. From this, one might therefore assume that the 1/1,00053 to 1/5,2248 risk rates reported in the 1999 and February 2000 reports are possibly due to multiple procedures: which are often a common plastic surgery practice. There is, however, no information as to whether patients who were the basis for this study had in fact undergone any other procedures in addition to liposuction.

The mortality for liposuction alone is likely much lower, but the risk may increase dramatically if other procedures, or perhaps general anesthesia, are involved.

Table 4 shows the incidence of liposuction complications in the literature. Some of this data comes from dermatologists' studies, which may on average reflects smaller volume liposuction. Dermatologists likely perform fewer combined and lengthy surgeries, and of course do not usually use general anesthesia. This is in contrast to the plastic surgeons' studies, which reflect all of these risks. Also, some complication rates cited are actually suspect. For example, the nausea/vomiting rate of 1/9869 in one plastic surgeons' study is certainly underreporting error, leading us to suspect other such problems in this paper.

Table 4. Liposuction Morbidity and Mortality Rates

Liposuction Morbidity

Complication Rate

Irregularity 1/4583
Paresthesias 1/41783
Superficial injury resulting from adhesive 1/48383
Seroma/Hematoma 1/19284*, 1/34554**, 1/61283, 1/23185
Persistent postoperative edema 1/83483
Allergic reaction to medication or adhesive 1/91883
Ultrasound-assisted liposuction skin burns 1/1,40469***
Excessive intra-operative or postoperative blood loss 1/1,83683, 1/6,71169, 1/82,707 fatality8
Infection 1/10785, 1/19284, 1/94854, 1/4,58983, 1/70,892 fatality8
Nausea/Vomiting 1/9869
Postoperative pneumonia 1/25584
Pulmonary embolism 1/76684, 1/3,75969, 1/8,39954, 1/9,01485, 1/16,541 fatal PE8
Major skin loss/Skin slough 1/76684, 1/1,10769, 1/15,11854
Deep venous thrombosis 1/76684, 1/3,04069, 1/3,02454
Risk of pulmonary embolus in patient with known deep vein thrombosis 1/12 - 1/1,42886
Compression foam blisters 1/7.784(estimated)
Mild skin necrosis/Minor skin loss 1/4584, 1/55054
Mild anemia 1/4384
Fluid overload 1/7784, 1/7,24669
Fat emboli 1/18,86869, 1/75,59154
Lidocaine toxicity 1/47,61969
Surgical shock 1/90,90969
Cannula penetration of abdominal cavity 1/37,79554, 1/47,61969, 1/26,118 fatal penetration8
Admission to hospital 1/90085 (2 infection, 1 phlebitis, 1 pulmonary embolis, 2 observation)
Death 1/5,0008, 1/37,79554, 1/41,66653, 1/47,41569

* In the study, 766 patients underwent large volume ultrasound-assisted liposuction.

** Major liposuction study (pre-tumescent technique) involving 75,591 patients.

*** Study involved 94,159 liposuction procedures. However, 14% of the research sample group included patients who underwent liposuction along with abdominoplasty.

Abdominoplasty: Worth the Risk?

"The patient contemplating an abdominoplasty and the surgeon planning it should be aware of the unpleasant reality of these complications, which range from the annoying to the lethal."87Abdominoplasty is acknowledged as posing the greatest risk of all cosmetic surgery procedures. Indeed, deaths and injuries caused by abdominoplasties far exceed those of not only other cosmetic procedures but also some other non-elective major abdominal surgeries.

The 1977 report published by Grazer and Goldwyn that surveyed 10,490 procedures calculated a death rate of 1/61788. Teimorian and Rogers55 showed a death rate of 1/2,415 from an even larger sample size of 26,562. A 2001 ASAPS survey in which records were analyzed from roughly 13, 000 patients who had undergone abdominoplasty between September 1998 and September 2000, calculated that the mortality rate amongst this group (who also had a liposuction procedure as an adjunct to their tummy tuck) was 1/3,28170. Refinements and improvements in many aspects of the technique itself and also in patient care and management during and following surgery could have improved the mortality rate for abdominoplasty almost five-fold in the intervening two decades since the 1977 study.

These reports would seem to the best we have so far in terms of standards from which to assess abdominoplasty risk rates. Most other studies have been based on patient sample numbers that are too small from which to calculate mortality rates, although they at least document morbidity rates and complication trends. Since some liposuction is performed with virtually every abdominoplasty, these most recently cited rates, as suggested by the authors, almost certainly reflect the underlying abdominoplasty mortality rate rather than an increased rate due to both procedures being performed together. Wound healing complication rates are especially high among abdominoplasty patients and the frequency of secondary corrective procedures and the increased lengths of stay in hospital required as a consequence are very high, occurring in roughly 1/3 abdominoplasties performed by plastic surgeons in these studies.

Pulmonary embolism, hematoma, dehiscences, abdominal perforation, infection, and seroma are other frequently documented complications8, 88-91. T-type plasty procedures appear to carry a particular risk of wound complications, especially necrosis, with secondary surgical correction being frequently required as a consequence. A French retrospective study of postoperative complications seen at long-term follow-up of 258 women following abdominal dermolipectomies, performed between January of 1991 and May of 1996, reported hemorrhage as a complication in 1/83 patients, lymphorrhea in 1/90 patients, infection in 1/14 patients, skin necrosis in 1/15 patients, secondary scarring dehiscence in 1/43 patients and thromboembolic events in 1/83 patients90. Infraumbilical plasty and full plasties with horizontal or inverted T scars, with our without lipoaspiration, were the surgical techniques employed. A significant difference in the rate of skin necrosis was observed between the T-type plasty (seen in 1/3 patients) and the other procedures (only 1/71 patients for infraumbilical plasties and 1/22 for full plasties with horizontal scar). Secondary corrective surgical procedures were required in as many as 1/3 patients90. Norwegian survey study of patients who had undergone abdominoplasty combined with closed liposuction of flap and flanks similarly documented that wide undermining and the "opposite T" incision emerged as significant risk factors related to flap necrosis, with a high proportion of patients developing this type of complication92.

A Dutch study has documented an increased rate of wound complications for male abdominoplasty patients with as many as 1/2 men (from a total of 101) reported as having suffered either wound infection, partial dehiscence, seroma, hematoma or skin edge necrosis92. In contrast, 1/7 female patients developed wound complications. A total of only 86 cases was analyzed retrospectively for this study; almost 1/10 of these documented an injury sustained to the lateral cutaneous nerve of the thigh92.

Another complication reported following abdominoplasty in males but not females is that of severe gastroeophogeal reflux, secondary to increased intra-abdominal pressure. The type of regional, male-specific obesity associated with this risk in abdominoplasty should be recognized preoperatively as it may place the patient at higher risk93. Unlike the massive increase in risk rates for other cosmetic procedures seen when abdominoplasty is performed as an adjunct - with liposuction in particular - the converse does not appear to be true, in that the inclusion of other procedures along with abdominoplasty does not appear to increase complication and mortality rates89, 90, 93. The obvious conclusion is that abdominoplasty is the independent variable: the high risks stem from this procedure rather than the others.

A factor that consistently demonstrates itself to be a significant risk in abdominoplasty is smoking. Complication rates for abdominoplasty are significantly higher in patients who smoke, have diabetes and/or hypertension90, 92, 94. Obviously, many surgical outcomes are adversely affected as a result of smoking, with intraoperative and postoperative pulmonary, cardiovascular, and cerebrovascular complications. However, the increase in wound complications is thought to be due to nicotine-induced vasoconstriction and consequent tissue ischemia that, with the added effect of wound-closure tension, is a particular issue of risk to optimal wound healing in abdominoplasty procedures.


Hematoma is the most frequently reported complication associated with rhytidectomy. The major precipitating factors are reported to be extensive undermining of flaps, straining due to coughing or vomiting, and high postoperative blood pressure95. The incidence of more clinically relevant major hematoma after facelift is one in every 45 patients96. Major hematomas require immediate attention to prevent the skin necrosis and infection that can develop as a consequence97. Hematomas generally form within the first 48 hours post-surgery, with major hematomas developing more rapidly, often within the first 12 hours following rhytidectomy surgery, and are more common in males due the increased blood flow in their thicker and more sebaceous skin97, 98.

These recent rhytidectomy-associated hematoma incidence rates agree roughly with an earlier study in which an overall risk of major hematoma in facelift patients were determined as being 1/5499. This study also went further to report on an association between hematoma risk and the level of anesthesia applied during the procedure. The incidence of hematoma in patients having general anesthesia was calculated as 1/90, whereas this rate increased to 1/27 in patients who had rhytidectomy under local anesthetic. Of those patients who had undergone the procedure with monitored intravenous sedation, only one per every 114 individuals developed hematoma98. Use of general anesthesia or monitored anesthesia usually allows more precise control of blood pressure than with local anesthesia, and many facelift surgeons including the author believe that low, carefully controlled blood pressure during the case contributes to less bleeding and possibly produces a lower hematoma rate.

Thrombosis, infection, seroma, necrosis, skin sloughing and alopecia are other, although somewhat poorly documented, complications of rhytidectomy96. Sensory nerve injury, particularly damage to the greater auricular nerve, is also common, leading to a loss of sensation of the lower ear area that often lasts for several weeks post-surgery in as many as one out of every 14 patients and may be permanent in some cases96, 99. Facial paralysis resulting from motor nerve damage during rhytidectomy is less common, but still occurs in 1/188 to 1/38 surgeries100. Of the nerves of the face, the frontal branch is most commonly injured, being more vulnerable due to its path over the zygomatic arch. The mandibular branch nerve is typically injured in the areas where it crosses the mandible as it courses toward the lateral mouth. There is also the much less commonly reported possibility of buccal injury, which can happen when deep dissection is carried out medially in the mid-face area96. The subperiosteal facelift appears to have a higher rate of all of those complications than standard subcutaneous facelifts100. Contemporary reports calculate that minor hematomas occur in one out of every 15 rhytidectomy patients101.

In terms of other serious complications and mortality associated with cosmetic facial surgery procedures, published risk rates for rhytidectomy are scant. A 1998 survey of plastic surgeons reported an incidence of deep vein thrombosis and pulmonary embolism following facelifts in one out of every 256 and 625 patients respectively101, 102. Ten to fifteen percent of clinically diagnosed pulmonary emboli are fatal103, 104, 105. Taken together, the data support an estimate of a fatality rate after facelift due solely to pulmonary embolis at 1/5000 or higher.

The conclusion of a more recent study in which one-third of active, randomly selected ASAPS members were surveyed in order to assess the incidence of, and factors associated with, deep venous thrombosis and pulmonary embolus after facelift during a 12-month study period was that deep venous thrombosis/pulmonary embolus after facelift is a significant complication of facelifts, one that had been seen by one out of every nine surgeons surveyed. A mortality rate of 1/9,937 was reported from the facelift procedures assessed in the one-year study period.105 Deep venous thrombosis occurred in one in every 286 patients and pulmonary embolus at a rate of 1/714. This survey study showed that deep venous thrombosis/pulmonary embolus is more likely to occur when the procedure is performed under general anesthesia, with a rate of 1/1.2 (83.7 percent) of deep venous thrombosis/pulmonary embolus events occurring in patients who underwent the procedure under general anesthesia. Intermittent compression devices were associated with significantly fewer thromboembolic complications from this report, with the authors suggesting that aesthetic surgeons consider adopting intermittent compression devices when performing facelift under general anesthesia106. Similarly, the conclusion of a retrospective study carried out by UCLA Head and Neck Surgery (Otolaryngology) of 114 facelifts was of a major complication frequency of 1/19 and minor complications in one out of every four patients106.

Blepharoplasty, Cosmetic Breast Implantation

For blepharoplasties, an increasingly popular procedure, retrobulbar hemorrhage and secondary blindness remain significant risks,107-109although statistics for just how many patients are affected are not documented.

Breast implantations are well known, with some authors110 stating that every cosmetic breast augmentation procedure is accompanied by some complication, be it capsular contraction, scarring, deflation, asymmetry, infection, hematoma, or other(s). A recent Danish study111 confirms this in part and provides some statistical insights into the complications associated with breast augmentation. They concluded that out of 1,572 implants performed over the period of the study, 1/6 was followed by one complication, 1/27 was followed by two complications, and 1/125 was followed by three or more complications. Capsular contracture was the most common complication, leading most frequently to additional surgery/hospitalization, and occurred in 1/9 of implantations. Hematoma occurred at a rate of 1/43 and infection in 1/50112.

Mortality data has heretofore been absent from the literature for cosmetic breast augmentation alone. This forces us to make possibly the crudest of approximations as to the number of deaths that arise from breast augmentation procedures, based on the following anecdotal reports. We do so mainly in the hope that we will ignite the spark that will drive others to perform the relevant investigative studies that are required to either refute or substantiate these figures.

In 2000-2002, in the Los Angeles area (excluding San Diego), we are aware of four breast implant surgeries involving mortality. Significantly, none of these fatalities were related to implant construction and some seem related to physician judgment (if you do not include prolonged general anesthetic operating time and multiple procedures). The author had no access to patient charts or first-hand reporting. The patient area covered roughly half the population of California. The ASPRS reported 206,354 U.S. breast implant procedures performed by plastic surgeons in 2001,112 which probably represents 95 percent of the total cosmetic breast implant surgeries113. The population of California is 34,000,000, and the U.S., 285,000,000114. If these were the only fatalities occurring with breast implant surgery in this region, and the above assumptions were correct, the mortality in Los Angeles over this time period can be calculated by:

34 million/2
285 million
  x 206,354 surgeries nationwide =  12,308 breast implant surgeries per year in Los Angeles


Based on this formula, we calculate that approximately 12, 308 breast implant surgeries are performed in Los Angeles area each year or 24,000 in two years. If four fatalities from this group were as a result of breast implant procedures, this equates to a 1/6,000 mortality rate: surprisingly close to the mortality from major studies of other cosmetic surgeries. If we assume typical underreporting bias, this figure could increase to 1/3,000.

Another way to extrapolate these figures is as follows. Mentor sells approximately 5,000 breast implants per annum in Los Angeles, Ventura, and San Bernardino Counties. These represent around 80 percent of the total market115 and from this we can crudely estimate that 7,000 breast implants are performed across this region each year. This would about cut in half the presumed denominator of the above fraction, which would double the mortality rate guess.

Previous page     Next page