PARIS +33 (0)1 49 23 00 81
GENÈVE +41 22 90 01 800
LONDRES +44 207 636 4272

Publication Transplant.


Long-Term Results of Fat Transplantation : Clinical and Histologic StudiesStockholm – Livingö, Sweden, and Tartu, Estonia

Nine cases of idiopathic constitutional subcutaneous fat atrophy in the cheek area during the years 1988-1992 were treated by the autologous transplantation of fat for purely aesthetic reasons. The fat graft suspension was obtained by the low-power aspiration technique with the use of a vacuum pump. The contour defects were initially overcorrected by approximately 50 percent more volume than required. The patients were followed up for 1.5 to 4.5 (mean, 3.5) years. Contrary to the experience of others, only the partial resorption of the transplanted fat occurred. This result was serified by six biopsy specimens obtained in the time range of 7 to 36 months after transplantation. The additional injections of fat were not necessary. Delicate tissue handling and the small total amount of fat transplanted by careful distribution in the recipient tissues are probably the factors responsible for the long-lasting improvement in these patients. As an experiment, suction of fat under -0.5 atm and -0.95 atm. was performed in five patients undergoing abdominal liposuction. Aspiration under maximum negative pressure caused partial breakage and vaporization of the fatty tissue. The diameter of the fat cells in the remaining beads of fat was in all five cases mechanically distended and thus was larger than in the lipocytes extracted at -0.5 atm. (Plast. Reconstr Surg. 94: 496, 1994.)
The fat cell is the best friend of the plastic surgeon. Free fatty tissue transplant to correct congenital or posttraumatic defects of the facial contours has been used occasionally for the last 100 years (1-4). An excellent historical review on the use of free autologous fat grafts was published by Billings and May 5, and a histologic review was conclucted by Smahel 6. At present, two fat survival theories exist that are based on microscopic examinations of the behavior of the fat autografts in the animal models. The host replacement theory postulates that the histiocytes would scavenge lipid material and eventually replace all adipose tissue. The more popular theory in recent years is the cell-survival theory, which states that some of the graft adipose tissue survives after the host reaction subsides (4-5).
In the last two decades the interest in the correction of soft-tissue defects has been centered on the various artificial tissue substitutes : silicon, collagen, and hydroxyapatite, for example. With the explosive gain in the popularity of liposuction, our attention recently turned toward the possibilities of the use of autologous fat transplants as a soft-tissue substitute (7-9). At first, the need to correct the dimples and waves that occur as undesirable side effects of liposuction procedures gave a new incentive to fill these defects with the original missing material-fat (8, 10,11). Encouraged by the results obtained, the procedure was used for a second main inclication, the correction of aesthetic contour deformities in the face (7, 12-14).
The transplantation technique underwent development from transplanting fatty tissue en bloc 24, to manually excised and prepared 4 to 6 mm large « pearls » of fat 8, and finally to fat tissue suspension obtained by manual power or vacuum machine-assisted suction (7, 9-11). With the latter technique we treated nine patients with idiopathic constitutional subcutaneous fat atrophy in the cheek area. The indication for operation was purely aesthetic in all cases.
During the last few years various reports have appeared about the use of fat obtainecl by vacuum aspiration for the improvevnent of the facial contour (12, 14-16). Many publications on the operative technique are available, but the reports on long-term results are scarce. Stimulated by the article by Goldwyn (17), we decided to investigate how long the implanted fat would last in patients treated because of the bilateral idiopathic constitutional fat atrophy in the cheek area. It was also interesting to test the hypothesis that the aspiration with the lower negative pressure should be used for the fatgraft harvesting (13).


Fat grafts were preferably obtained from localized fat deposit areas with a low degree of vascularization, for example, the trochanteric area or upper part of the inner thigh to decrease with extravasated blood and thus diminish the risk of infection. Fat in the periumbilical area is more richly vascularized thus less suitable, but it was the only donor site available in the thin patients.
Both parts of the procedure were performed in the same session, and we used dissociative and local anesthesia. State of dissociation was obtained by medication with kétamine and diazepam according to the schedule outlined by Gordon (18). Lidocaine in 1% and 0.5% solution was used for the local infiltrative anesthesia. Adrenaline was excluded because of the fear that it would jeopardize the adipocytes. Excessive infiltration of the donor site with the normal saline solution was avoided because this disrupts agglomerations of adipocytes. The so-called dry technique was used (11).
The fat tissue was extracted with the help of cannulas of 3, 4 or 6 mm diameter. The cannulas were connected through plastic tubing to the sterile fat collector of glass, which in turn was connected to the vacuum pump. We also used low-power suction with negative pressure of -0.5 atm, believing that it would decrease the number of ruptured adipocytes (13).
The aspirated fat tissue was handled delicately to prevent mechanical injury to the fat cells. A strong asepsis was observed througout the whole procedure. The aspirated was rinsed with normal saline solution to remove the blood component. This was done by filling half or less of a 50 ml syringe with the fat tissue through the piston opening, reassembling the syringe, aspirating the normal saline solution, and turning the syringe up and down a few times. After that the syringe was placed in the vertical position. The saline and blood descend downward, and the fat as a lighter substance floats up, with the triglycerides released from the damaged lipocytes floating on the top. Usually three of such washings are required, after which the fat has a clean yellow ? white color and is ready for transplantation.
The borders of the receiving area were marked with methylene blue, and infiltration with the local anesthetic performed subcutaneously in the periphery of the area ; only inside the marked circle was it deep, and that was mostly underlying musculature.
The specially made postol (MD ingineering, Hayward, Calif.), which apportions the injected fat in the range of 0.5 to 1.0 ml, was attached to the injecting cannula. This cannula has an inner diameter of 2.3 cm and a rounded and flat tip, which is suitable for the blunt dissection. The distal opening of the cannula faces downward during the procedure. Cannulas or needles much narrower than that cause breaage of fat cells. It should be noted, however, that the conical outlet of the tip of the syringe has the inner diameter of only 1:8 mm.
An awl was used to make a stab wound incision outside the depressed area, and the incision was usually hidden in the sideburn, in the natural wrinkle near the ear, or on the inside of the lips commissure. Such puncture wounds do not require suturing. At first the cannula was gently pushed insicle the buccinator and the anterior part of masseteric muscles, and deposits of fat were made in several locations. We then proceeded into the localization corresponding to the fat pad of Bichat and finally to the more superficial plane between the muscles and subcutis. The surgeon’s left hand was on the cheek the whole time, controlling the position of the cannula. Transplanted fat was apportioned and placed according to individual requirements. Usually more fat was transplanted centrally, with a gradual decrease of augumentation toward the periphery of the deformity. The amount of transplanted fat to each site varied between 7 and 27 ml (mean, 18 ml). The contour defects were were initially overcorrected by approximately 50 percent more than desired augmentation.
The patients were given facial garment support for the first postoperative week to reduce the edema and to hide the initial overcorrection. They were advised to massage the treated area with their fingertips for a few minutes four times a day during the first 5 days after the procedure. The massage helps to equalize distribution of the transplanted tissue. Both local heating and cooling were discouraged.

The Follow-up Study

The studied group consisted of eight women and one man. They were treated by the free autologous transplantation of fat, the procedures were carried out during the years 1988 – 1992. In all patients both the left and the right cheek were augumented. Their ages ranged from 42 to 65 (mean, 48 years). All patients had good general health and, in varying degrees, neurotic personalities. Two were extremely thin, and seven had a normal constitution. Their deformity, atrophy or absence of the natural subcutaneous fat in the cheek region, was graded as severe in five and moderate in four cases, depending on the size of the defect (irregularity) in the cheek contour.
Control examinations were made at least once or twice a year. Color photoprints were obtained in the standard projections ; frontal, three-quarter, and lateral views were used as an aid to objectivize recorcled results. The length of follow?up was 4.5 years in three patients, 3 years in two patients ; and 4, 3.5, 2.5, and 1.5 years in one patient each (Table I). The mean follow-up time was 3.3 years.
Six biopsy specimens of fat from the transplanted areas were obtained 7 to 36 months after implantation. Three of our patients had biopsies once, and the fourth had biopsies three times. Five biopsy specimens were obtained under direct vision 7, 10, 14, and 25 months after transplantation and in one instance by liposuction 36 months after augumentation. In one patient, E.S., a futile attempt was made to correct a clark ring under the left eye by transplantation of fat, the attempt was performed at the same time as the cheek augumentation. This fat was placed below the subcutis and on the orbicularis muscle. It survived, but 7 months later had to be removed because of the bad cosmetics. Five other biopsy specimens were obtained from the deep subcutis on the cheeks, from the locations corresponding the center of transplanted areas.


Case Summaries of Nine Patients with Moderate or Severe Subcutaneous Fat Atrophy
in the Cheeks Treated by the Autologous Transplantation of Fat.

Patient Grade of Deformity Transplanted Volume (ml) Length of
Biopsy Elapsed Time
2 x 20
2 x 15
2 x 20
9 x 27
2 x 20
2 x 14
2 x 19
2 x 7
2 x 22


7 months
14 months
10 m. 25 m. 3 y.
21 months

Note : Only the partial resorption of the transplanted fat occurred.
Mean length of observation was 3.5 years.

The specimens were fixed in Zenker’s solution for paraffin sections, which were subsequently stained with hematoxylin and eosin. Calculations of the size of adipocytes for the assessment of the fat cells after various periods of time after implantation were made by the method of direct microscopy. The cliameters of 50 to 100 fat cells were measured under the 100-power objective of a binocular light microscope with grids in the 10-power eyepiece. Length and width was measured in the each cell to determine mean diameter. The average calculation for each biopsy specimen was made with t test as a statistical method.

The Experiment

The specimens for differential pressure studies were obtained from the five patients scheduled for the abdominal liposuction. Before this procedure, the biopsy specimens of fat were aspirated from two « virginal » areas, from one to the left of umbilicus with -0.5 atm and from another to the right of umbilicus with -0.95 atm. The 6 mm abortion cannula with the single-side opening was used. The biopsy specimens were collected directly from the distal part of the hose.
The viability index for adipocytes aspirated under the various negative pressures was estimated by the direct microscopy of the fresh specimens with a hemocytometer-Bürken’s chamber. For this purpose fat cells were treated vith trypsin and stained with trypan blue. Calculations of the size of adipocytes for the assessment of influence of strength of the negative pressure were made with the same methods as used for the biopsy specimens in the follow-up study. Paired t test was used to evaluate differences in size of the adipocytes. The other part of the aspirate was fixed in Zenker’s solution and stained with hematoxylin and eosin.

Clinical Observations

One patient maintained improved contour of her cheeks for 1½ years. Then, she looked elsewhere to have her fat grafts suppressed by triamcinolone acetonide injections. The rest of the group were satisfied with the change of contours of their cheeks. Contrary to the experiences of Illouz (11), Ersek (19), and Ellenbogen, at the present time only partial resorption of the transplanted fat occurred in our patients. The clinical appearance of each of them and their photographs were evaluated. Precise volumetric analysis of the treated areas was not possible, but our general impression is the same in all nine patients. They retained approximately 40 percent to 50 percent of the initially transplanted volume. The additional transplantation of fat was neither requested nor judged necessary.
Three patients had stable cheek contours at the end of the study. In one patient the improvement started to fade away during the fourth year after transplantation. Four patients gained weight (3, 8, 8, and 10 kg, respectively) more than 1 year after fat transplantation. Remarkably, we observed that transplanted areas not only increased in volume but also that they seemed to exceed the average increase of tegument in the other areas of the body. The male patient had such excessive bulging of his cheeks that he requested and underwent liposuction of the previously augumented areas 36 months after the transplantation. A total of 12 ml of fat was removed from the right and 10 ml from the left cheek. The specimens were, of course, examined histologically.
The only complication in this series was granuloma formation in one cheek, which was certainly caused by too large an agglomeration of the fat graft in one location. It was cured by intraoral incision, curettage, and antibiotics. The overall cheek contour was not disturbed.

Microscopic Observations in the Follow-up Study

The diameters of fat cells calculated from the biopsy specimnens obtained from the transplanted areas are shown in Table II. Histologic analysis showed that already after 7 months the transplanted fat had an organized lobular structure containing rather small adipocytes. Their size was in range of 10 to 70 µm. Each lobulus contained several hundred or thousand viable fat cells and contained its own vascular net.
In the more long-lived specimens we saw pronounced fibrosis between lobuli and sometimes slight intralobular connective tissue formations. This lobular structure was well preserved in the specimens removed under direct vision at the face-lift operations, but it was much more deranged in the specimens obtained by the blunt aspiration.
In the patient in whom the augumented cheeks were corrected by liposuction, enlargement of the cheek volume corresponded to an increase in the average size of the fat cells, as seen at microscopy. Most of them measured 60 to 80 µm. There was no doubt about the viability of the fatty tissue in all six specimens. We did not find any signs of fat necrosis, such as severe atrophy and breakage of large numbers of adipocytes or oil cysts.

Microscopic Observations in the Experiment

In five patients the viability of adipocytes was compared in the fat beads obtained by liposuction under the maximum negative pressure of -0.95 atm and under the lower negative pressure of -0.5 atm. Fat aspiration with the maximum negative pressure cause partial breakage and vaporization of the fatty tissue. This was seen seen as the opalescent film inside the transparent connective tube. Another part of the pat was smashed against the walls of the collector, breaking the fragile walls of adipocytes. Liberated triglycerides and free lipides were accumulated with the fat beads in the collector, but they could be decanted.
The remaining of fat were examined. No macroscopic difference was noted, as compared with the fat extracted under -0.5 atm. Direct light microscopy with the hemocytometer was applied for the viable and damaged cell count. We estimated that about two-thirds of the fat aspirated under -0.95 atm pressure withstood the trauma of extraction. In this part of the aspirate, more than 90 percent of adipocytes had their walIs intact.
Statistical analysis of the size of adipocytes aspirated under -0.5 and -0.95 atm negative pressure was conducted by comparing means of each hypothetical population of five patients. A statistically significant (p < 0.001) decrease of the cell size was noted in the group exposed to the lower (-0.5 atm) negative pressure. Adipocytes in this group were 29 percent smaller than those aspirated under -0.95 atm.


The technique of autologous transplantation of fat is not new, but until the last few years the results were unpredictable because of the lack of knowledge of the proper technique and indications 19. Therefore, most surgeons were and still are skeptical. Illouz 11 pointed out the phases of the technique of free autologous fat transplantation, each of them influencing the take of the graft or rejection : choice of donor site, method of fat removal, method of preparation and transplantation, selection of the recipient site, and preparation of the recipient area. The rationale of transplanting a limited amount of fat and placing it in contact with the welle vascularized tissue is confirmed by the histologic studies of Peer 4, Carpaneda and Ribeiro, and the present report.


Diameters of Fat Cells calculated from Biopsy specimens
obtained 7 to 36 months after autotransplantation
Time after Implantation
Patient Diameter of
fat cells (µm)
7 E.S. 50.6 ± 1.2
10 Z.P. 60.1 ± 1.4
14 M.W. 71.4 ± 1.9
21 D.N. 52.6 ± 1.1
25 Z.P. 55.2 ± 1.3
36 Z.P. 67.9 ± 2.4

The transplanted fat is revascularized in the centripetal manner. Its capacity to survive through the plasmatic imbibition from the edges is 0.5 to 1.5 mm. Thus, the single threads of the transplanted adipose tissue should not exceed 3 mm in diameter. The larger single collections of fat are subject to liquification, central necrosis and cyst formation. This, in addition to technical imperfections, was the reason for reports of remarkable resorption rates and the general unpredictability of the results (10, 19).
The choice of the donor site can be a problem because many of the patients seeking augmentation of the facial contour are very thin. In general the sites with soft and easily removable fat should be used : lower abdomen, upper inner thigh, or trochanteric area. Hudson et al. identified cell size, lipogenic activity, and the presence of aclipose tissue lipoprotein lipase as the factors promoting revascularization and increased viability of the graft. They found that the fat in the femoral region is more suitable for transplantation than facial fat.
Fat tissue in the lower half of the body has less coonective tissue septa and thus is only sparsely vascularized. This decreases the amount of blood in suctioned fat. Extravased blood is known to be a good media for bacterial growth.
The technique presented here or single transplantation with overcorrection was worked out by the senior author (Niechajev) by reading the literature, listening to the others, experimenting, and testing empirically. As the logical consequence, we observed the principles of strict sterility, apsiration of fat with low negative pressure, gentle tissue manipulation, rinsing of fat in normal saline solution, injecting through the large (2.3 mm) cannula the flat blunt tip performed in the crisscross manner and in several layers, transplanting only on the withdrawal movements, and keeping, and keeping within 15 to 20 ml per site. Thus, we came independently to the same conclusions as the Chajchir et al. 15, Chajchir and Benzaquen, and Matsudo and Toledo who used the meticulous technique ans sterility similar to ours. Coiffman, on the other hand, advocates undercorrection but with several transplanting sessions at intervals of at least 1 month.
Our differential negative pressure studies confirme the rationale of using low negative pressure for extraction of the fat grafts. We assume that the mechanical distention of adipocytes increases the risk of and sometimes cause cell breakage.
Adipocytes in biopsy specimens obtained from the sites of transplantation were generally of small size (Table II). The size varied from 50.6 to 71.4 µm, which we interpret as simply individual variations. As a reference we used the calculations of Hudson et al., who determined the average diameter of the adipocytes. They were largest in the femoral area, 104 µm, and gradually decreased in the higher parts of the body, being smallest in the face (81 µm). We can therefore see the general tendency of some shrinkage in the transplanted fat celIs in our material.
How do we know that we biopsied a grafted fatty tissue and not a normal one? The proof was definite in one patient, but only circumstantial in the other three cases. Despite many doubts the achievement of our clinical goal – the long-term manitenance of the increased cheek volume – cannot be disputed.
Our clinical and histologic examination proved that equal, multilayered distribution of narrow strings of fat promoted the revascularization process of fat without the stage of cystosteatonecrosis. The chance of healing was good because of the favorable volume / surface ratio. The results were more stable than those seen with collagen. We have found that fat absorption slows down after approximately 2 years, and (usually) after that, 40 percent to 50 percent of the original volume of fat is retained. That fat is revascularized and is susceptible to the voumes changes after several weight gain by the patient. This confirme the earlier observations of Peer (4) that free autogenous fat grafts may actually increase n size if the patient takes on an increase in adiposity that affects the particular fat system from which the graft was taken.
Why are our results more promising than the others ? The transplanted fat survived in our patients as a fat without extensive fibrosis observed in the material of Chajchir and Benzaquen. They usually transplanted larger volumes per site (up to 180 ml per cheek). We used only the relatively small amounts of fat (on average 18 ml per cheek).
Our findings give further evidence for support of the cell survival theory in free fat autotransplantation. The graft of adipose tissue goes through an initial period inadequate nutrition through plasmatic imbibition. Earlier histologic investigations on the behavior of the fat grafts in human beings are sparse and were made days or a few months after implantation. Our biopsy specimens were obtained many months or 2 to 3 years after transplantation. We can therefore not answer the following intriguing question : did the surviving adipocyte undergo transformation by dedifferentiation to the preadipocyte stage and differentiation again into the mature adipocytes, or did they survive by revascularization of the vascular net in the connective tissue stroma of the fat ? It is plausible, however, that fat cells present a few years after transplantation are surviving or descendent fat cells from the original graft, as Peer and Smahel (personal communication) have noted.
How does fat grafting into the cheek area compare with the other recipient sites ? We share with Matsudo and Toledo the favorable experience in the autotransplantation of fat for the correction of skin dimples as a sequela of liposuction and the excellent experience of Wilkinson in the use of the fat graft for lip augumentation.
We raise a warning regarding the use of lipoinjection for the correction of the glabellar frown lines and the periorbital area. We know of four cases of unilateral blindness and one case of severre damage to the central nervous system, all of which occurred after fat injection under the glabellar frown lines. The probable cause was fat embolism in the central retinal artery and in the cerebral arteries, respectively.
We conclude, similarly to Nguyen et al, that fat grows on fat and that it does even better when placed intramuscularly. What stays after 2 years is viable ans permanent. Fat transplanted on ther surface of the bones, tendons, fascias and aponeuroses, as a rule, completely resorbs. Autologous transplantation of fat undertaken in the cavalier manner, without knowledge of the proper technique and inclications, gives at best unpredictable results.
Igor NIECHAJEV, M.D., D.Sc., and Oleg SEVCUK, M.D


  • Neuber, G. A. Fettransplantation. Chir. Kongr Verhandl. Deutsch. Gesellsch. Chir. 22: 66, 1893.
  • Moszkowicz, L. Fettplastik bei Hemiatrophia Faciei. Med. Klin. 26: 1478, 1930.
  • Eitner, E. Fettplastik bei Gesichtsatrophie. Med. Klin. 27: 624, 1931.
  • Peer, L. A. Transplantation of Tissues, Vol. 2. Baltimore: Williams and Wilkins, 1955. Pp. 195-230.
  • Billings, E..Jr., and May J. W. Jr. Historical review and present status of free fat graft autotransplantation in plastic and reconstructive surgery. Plast. Reconstr. Surg. 83: 368, 1989.
  • Smahel, J. Adipose tissue in plastic surgery. Ann. Plast. Surg. 16: 444, 1986.
  • Chajchir, A., and Benzaquen, I. Liposuction fat grafts in face wrinkles and hemifacial atrophy. Aesth. Plast. Surg. 10: 115, 1986.
  • Ellenbogen, R. Free autologous pearl fat grafts in the face a preliminary report of rediscovered technique. Ann. Plast. Surg. 16: 179 1986.
  • Newman, J. Preliminary report on « fat recycling » liposuction fat transfer for facial defects. Am. J. Cosmetic Surg. 3: 67, 1986.
  • Bricoll, M. Autologous fat tissue augumentation. Am. J. Cosmetic Surg. 4: 141, 1987.
  • Illouz, Y. G. Present results of fat injection. Aesth. Plast. Surg. 12: 175, 1988.
  • Moscona, R., Ullman, Y. Har-Shai, Y., and Hirschowitz, B. Free-fat injections for the correction of hemifacial atrophy. Plast. Reconstr: Surg. 84: 001, 1989.
  • Chajchir, A., and Benzaquen, I. Fat-grafting injection for soft tissue augumentation. Plast. Reconstr: Surg. 84: 921, 1989.
  • Niechajev, I. AutologousTransplantation of Fat (Lipo-Filling) for the Improvement of the Cheek Contour : Long-Term Results. In U. Hinderer (Ed.), Plastic Surgery 1992, Vol. 2. Amsterdam : Excerpta Medica, 1992. Pp. 747-748.
  • Chajchir, J., Benzaquen, I., Wexler, E., and Arellano, H. Fat injection. Aesth. Plast. Surg. 14: 127, 1990.
  • Teimourian, B., Chajchir; A., Gotkin, R., and Reisin, J. H. Semiliquid autologous fat transplantation. Adv. Plast. Reconstr. Surg 5: 57-84, 1989.
  • Goldwyn, R. M. Unproven treatment: Whose benefit, whose responsibility: (Editorial). Plast. Reconst. Surg 81: 946, 1988.
  • Gordon, H. L. The selection of drugs in office surgery. Clin. Plast. Surg. 10: 277, 1983.
  • Ersek, R. A. Transplantation of purified autogenous fat