Prevalence, Severity, and Risk Variables Associated With Gingival ...

J Periodontol • June 2006

Prevalence, Severity, and Risk Variables Associated With Gingival Overgrowth in Renal Transplant Subjects Treated Under Tacrolimus or Cyclosporin Regimens Fernando de Oliveira Costa,* Se´rgio Diniz Ferreira,* Luı´s Ota´vio de Miranda Cota,* Jose´ Eusta´quio da Costa,* and Marco Antonio Aguiar*

Background: Some reports suggest that the prevalence and severity of gingival overgrowth (GO) is lower in patients taking tacrolimus (Tcr) than in those taking cyclosporin A (CsA). The present study was conducted to determine the prevalence and severity of GO in a group of renal transplant recipients treated with Tcr in comparison to those treated with CsA and to evaluate the risk variables associated with the development of GO in these two drug regimens. Methods: A cross-sectional study was conducted in a public hospital in Belo Horizonte City, Brazil. Demographic, pharmacological, and periodontal data were recorded for each subject. Variables from 134 subjects taking Tcr were compared to 451 subjects taking CsA using independent sample t, x2 statistic, or MannWhitney tests. The effects of potential risk factors on GO severity were determined using forward and backward stepwise regression analysis. Results: Subjects taking CsA showed a higher mean GO score (29.03% – 22.9%) compared to subjects taking Tcr (16.9% – 3.4%) (P = 0.0038). In the Tcr group, 17.9% of the subjects had clinically significant GO compared to 38.1% in the CsA group (P = 0.045). In the multivariate final model, papillary bleeding index, azathioprine dosage, and concomitant use of calcium channel blockers (CCB) were significant variables associated with severity of GO in Tcr and CsA groups. In addition, previous CsA use also remained significant for GO in subjects under a Tcr regimen. Conclusions: The prevalence and severity of GO is lower in transplant subjects taking Tcr compared to CsA. GO severity in both groups was strongly associated with the papillary bleeding index, highlighting the role of inflammation in this condition. Concomitant CCB use, azathioprine dosage, and previous CsA use in the Tcr group reinforces the possible synergistic effects of these pharmacological variables on GO severity. J Periodontol 2006; 77:969-975. KEY WORDS Cyclosporin; gingival overgrowth; prevalence; tacrolimus; transplants.

G

ingival enlargement can stem from a wide variety of causes. It is a significant condition that leads to functional, esthetic, and phonetic complications. Gingival overgrowth (GO) is commonly present in transplant recipients, normally induced by drugs such as diltiazen, cyclosporin A (CsA),1,2 and tacrolimus (Tcr).3 The management of drug-induced GO in this group of patients is a concern in the dental profession. CsA has been used in individuals who had received solid organ transplants in immunosuppressive regimens and is most commonly associated with other immunosuppressants such as azathioprine and prednisolone. Prevalence studies have shown that approximately 25% to 81% of patients medicated with CsA, with or without drug associations, experienced GO at some level.1,4 Tcr was introduced in 1987 as an immunosuppressive agent.3 The pharmacodynamics of Tcr are very similar to those of CsA, inhibiting calcium-dependent events, and are recognized as an effective alternative to the CsA regimen. In primary therapy, it has been suggested that Tcr may be even more effective than

* Department of Periodontology, Federal University of Minas Gerais, Belo Horizonte, Brazil. doi: 10.1902/jop.2006.050327

969

Prevalence and Risk Factors for Gingival Overgrowth

CsA in preventing acute and chronic rejection of liver transplants.3 Initially, some case reports have suggested that GO observed in subjects taking Tcr is lower than in those taking CsA, and the GO tends to be less severe.5,6 A high proportion of organ transplant recipients are concomitantly medicated with calcium channel blockers (CCB), usually nifedipine, to ameliorate the nephrotoxicity of the immunosuppressive agent.7 Concomitant medication with CCB and CsA or Tcr has, in previous studies, shown an increase in both prevalence and severity of GO.1,3 Some articles have reported that the GO in subjects taking Tcr appears to be more severe when CCB are also taken.3,8,9 The data of GO prevalence in CsA regimens has been extensively described in previous dental literature.1,10,11 Nevertheless, Tcr prevalence data and comparison studies among immunosuppressive regimens are limited.3 In addition, the relationships between GO and different pharmacological, demographic, and periodontal variables are still unclear.1,12 The purpose of this study was to determine the prevalence and severity of GO in renal transplant subjects under Tcr or CsA regimens and to investigate the effects of potential risk variables associated with GO on these groups of subjects. MATERIALS AND METHODS The total study population was 451 subjects under CsA regimen and 134 subjects medicated with Tcr, enrolled from January 2004 to April 2005. All renal transplant subjects were recruited from the organ transplant unit at Felıćio Rocho Hospital, Belo Horizonte City, Brazil, where they were seen on a regular basis to monitor drug therapy and graft survival. All subjects had transplants at least 2 months prior to the study, were at least 18 years old at the time of examination, and were receiving immunosuppressive therapy with either Tcr or CsA in combination with a range of other drugs. The inclusion criteria consisted of a minimum of six of the 12 most anterior teeth in the upper or lower dental arches. During the 6-month period of data collection, patients meeting the inclusion criteria were randomly invited to participate in the study according to the accessibility and availability of the subjects in the hospital unit routine. GO was assessed in relation to the following variables: age, gender, plaque score, papillary bleeding index, time since transplant, prednisolone and azathioprine dosage, and concomitant CCB use. The present study was approved by the Felıćio Rocho Hospital’s Ethics Committee, and an informed written consent was obtained from each subject. Periodontal Examination Gingival assessment was undertaken by one trained and calibrated periodontist who was blinded to each 970

Volume 77 • Number 6

patient’s identity and medical history. Calibration was done in the beginning of the study and repeated 1 month later. To determine intraexaminer reliability, 10 subjects were reexamined for papillary bleeding index and GO score in both Tcr and CsA groups. All k scores were >0.93, and intraclass correlation coefficients were ‡0.85. The level of oral hygiene was assessed using the plaque index.13 Measurements were made on the lingual, labial, and interproximal surfaces of the six most anterior teeth in the upper and lower arches. Scores for each site were summed and the mean values assigned. The papillary bleeding index14 was recorded in the interproximal area of the most anterior teeth. Upper and lower anterior teeth were assessed through visual inspection to record GO, and GO scores and the cutoff point for clinically significant GO were based on previous studies.4,10,15 The buccal and lingual papillae, associated with the six most anterior upper and lower teeth, were each assigned a score between 0 and 5, depending on the amount of both horizontal and vertical enlargement. Likewise, a total of 20 such papillae were selected and examined, presenting a potential maximum GO score of 100, which could be expressed as a percentage. Subjects with overgrowth scores ‡30 were classified as having clinically significant overgrowth.10 Pharmacological Variables Pharmacological data were obtained from each subject’s medical records. As part of long-term management, transplant recipients were screened regularly for whole blood and serum through concentrations of Tcr or CsA. Data from the most recent assessment, usually on the last medical examination, were recorded and used in the analysis. Patients’ medical records were thoroughly examined, and data were confirmed by the organ transplant medical group. Statistical Analysis Individual patient data, including demographic, pharmacological, and periodontal variables, were collected and transcribed into a statistical database.† The comparison of variables among the groups was performed using the unpaired independent sample t, x2 statistic, or Mann-Whitney tests, as appropriate. Statistical data analysis accounted for sample size differences of the two groups. Data were initially examined using univariate regression analyses to evaluate the effect of each independent variable on the prevalence and severity of GO, in both Tcr and CsA groups. The effects of the variables on the GO scores were subsequently examined using both backward and forward stepwise regression analyses and the general † Statistical Package for Social Sciences, Version 9.0 for Windows, SPSS, Chicago, IL.

Costa, Ferreira, Cota, Costa, Aguiar


linear model. The regression coefficients, their 95% confidence intervals (CI), and P values were also reported. RESULTS Demographic, pharmacological, and periodontal data from both groups were compared between the CsA group (N = 451) and Tcr group (N = 134). The CsA group presented more severe GO scores (29.03% – 22.9%) compared to the Tcr group (16.9% – 3.4%) (P = 0.0038). Subjects with GO under CsA regimen presented a higher prevalence of clinically significant GO (38.1%) compared to the Tcr group (17.9%) (P = 0.045). The periodontal variables (plaque and papillary bleeding index) were not statistically different between the two groups (P >0.05) (Table 1). Demographic, pharmacological, and periodontal data were compared for subjects taking CsA and Tcr and dichotomized into those taking CCB or not. The groups showed no statistical differences in the occurrence of the variables evaluated, except for the mean GO score and the percentage of clinically significant GO, which was greater for both CsA (P = 0.002) and Tcr (P = 0.001) groups when subjects were concomitantly medicated with CCB. However, it is important to emphasize that 10.4% of the subjects medicated with only Tcr and 34.5% of those taking only CsA showed clinically significant GO (Table 2). Univariate regression modeling in the CsA and Tcr groups revealed the following independent variables

associated with GO severity: age, papillary bleeding index, plaque index, medication with CCB, and azathioprine dosage (P 0.05). In addition, previous medication with CsA was also significantly associated with GO severity in the Tcr group (P = 0.045) (Tables 3 and 4). The stepwise regression and general linear models were used to identify independent variables associated with the severity of GO. Both multivariate final models for the Tcr group (adjusted R2 = 0.59%) and for the CsA group (adjusted R2 = 0.44%) failed to retain the plaque index and age from the previous univariate analysis. Azathioprine dosage, papillary bleeding index, and medication with CCB were maintained in the final model as risk variables associated with GO in both groups. Previous CsA use remained significant for GO in subjects under a Tcr regimen in the multivariate final model (Table 5). Moreover, upon evaluating the pharmacological data of the 24 subjects from the Tcr group with clinically significant GO, it was verified that 15 of these subjects had a history of previous CsA use (62.5%). Including previous CsA use in the multivariate regression analysis for the Tcr group, the adjusted R2 obtained was 0.59. When this variable was removed from the model, the adjusted R2 was 0.51, indicating that previous CsA use reveals an 8% increase in the GO score variations. CCB medication alone

Table 1.

Demographic, Pharmacological, and Periodontal Variables of Transplant Subjects According to Main Immunosuppressive Agent CsA (N = 451)

Tcr (N = 134)

% male (n)

59.7 (269)

53 (71)

% female (n)

40.3 (182)

47 (63)

Age (years; mean – SD [range]) Plaque score (range)

P

0.0621

47.6 – 8.4 (18-61)

42.3 – 8.9 (18-58)

0.0074

0.72 (0-1.73)

0.69 (0-2.43)

0.1742

Papillary bleeding index (range)

0.412 (1-3.7)

0.261 (0-1.74)

0.1773

Mean % GO score (– SD)

29.03 (– 22.9)

16.9 (– 3.4)

0.0038

52.8 (2-242)

42.0 (2-128)

0.279

6.5 (0-90)

6.0 (0-62)

0.058

50.0 (0-500)

50.0 (0-150)

0.730

57 (257)

42.5 (57)

0.565

38.1 (172)

17.9 (24)

0.045

Mean time since transplant (months [range]) Daily dosage of prednisolone (mg [range]) Mean daily dosage of azathioprine (mg [range]) % subjects taking CCB (n) % subjects with clinically significant GO (n)

971



Table 2.

Demographic, Pharmacological, and Periodontal Variables of Transplant Subjects Immunosuppressed With CsA and Tcr According to Concomitant CCB Use CsA (n = 194)

CsA + CCB (n = 257)

P

Tcr (n = 77)

Tcr + CCB (n = 57)

P

Mean age (years)

44.2

49.1

0.171

48.5

44.0

0.192

Months since transplant

53.1

51.2

0.698

42.0

39.0

0.745

Papillary bleeding index

0.38

0.44

0.160

0.189

0.293

0.347

Plaque index

0.72

0.83

0.182

0.56

0.832

0.575

Daily dosage of azathioprine (mg)

60.0

50.0

0.921

65.0

50.0

0.937

Daily dosage of prednisolone (mg)

10.5

12.5

0.522

7.0

8.5

0.430

Serum concentration (mg/l), main immunosuppressive agent

183.0

179.5

0.584

12.2

9.5

0.167

Dosage (mg), main immunosuppressive agent

211.5

200.0

0.475

7.0

6.0

0.520

Mean GO score (%)

23.5

31.4

0.002

9.96

24.86

0.001

% with clinically significant GO (n)

34.5 (67)

44.7 (115)

0.003

10.4 (8)

28.0 (16)

0.002

accounted for 29% of the GO variation (R2 = 0.29) in the Tcr group and for 27% (R2 = 0.27) in the CsA group. DISCUSSION A large portion of transplant subjects that take immunosuppressive drugs present significant functional, esthetic, and phonetic problems due to GO. In addition to this immunosuppressive treatment, a

large proportion of these subjects also take CCB, which can also induce GO. Some reports suggest that a synergistic effect stems from this type of medication.1-3,9,16 Overall, it has been a difficult task for medical personnel to change the immunosuppressive regimen to avoid this oral side effect. Few reports have evaluated the effects of Tcr in gingival enlargement. Studies failed to identify GO in subjects treated with this drug8,9 and others pointed

Table 3.

Univariate Analysis of CsA Group

972

95% CI for Slope

Adjusted R2 (%)

Regression Coefficient

P for Slope

SE for Slope

Gender

-5.21

0.097

3.85

Age

-0.426

0.031

0.185

-0.779

-0.049


-0.051

0.573

0.072

-0.18

0.16

2.08


20.80

0.042

8.19

6.19

39.43

23.48

Plaque index

12.47

0.038

3.94

0.509

23.69

10.68

CsA dosage

0.96

0.111

0.73

-0.465

2.62

2.72

Serum CsA concentration

-0.516

0.474

0.579

-1.79

0.598

0.719

CCB use

14.62

0.000

3.83

6.98

Prednisolone dosage

0.25

0.180

0.152

-0.071

0.562

3.31

Azathioprine dosage

-0.080

0.015

0.042

-0.177

-0.017

12.26

-14.69

3.18

21.88

3.12 12.3

27.16



Table 4.

Univariate Analysis of Tcr Group Regression Coefficient

P for Slope

SE for Slope

Gender

-4.73

0.095

3.19

Age

-0.382

0.028

0.166


-0.047

0.591

0.083


16.99

0.010

Plaque index

10.68

Previous CsA use Tcr dosage

95% CI for Slope -12.33

Adjusted R2 (%)

2.64

2.94

-0.701

-0.044

11.26

-0.22

0.11

1.95

6.42

4.75

28.91

16.91

0.046

4.91

0.622

20.38

10.20

8.73

0.045

3.42

-0.275

14.62

5.87

0.86

0.145

0.61

-0.369

2.13

2.32

Serum Tcr concentration

-0.424

0.392

0.519

-1.48

0.629

0.815

CCB use

14.71

0.000

3.21

6.87

Prednisolone dosage

0.22

0.122

0.148

-0.069

0.546

3.07

Azathioprine dosage

-0.082

0.012

0.036

-0.167

-0.018

11.93

Table 5.

Stepwise Regression Analysis for Tcr and CsA Groups Regression Coefficient

95% CI for Regression Coefficient

P

Tcr group Azathioprine dosage Previous CsA use Papillary bleeding index CCB use Constant

-0.0704 9.47 11.99 15.16 5.68

-0.14 3.04 2.30 8.74 -0.47

-0.000 14.89 21.14 20.09 13.33

0.050 0.046 0.019 0.0001 0.065

CsA group Azathioprine dosage Papillary bleeding index CCB use Constant

-0.0692 12.78 15.90 5.95

-0.13 2.59 8.85 -0.52

-0.000 21.67 22.19 12.60

0.046 0.014 0.0001 0.063

Adjusted R2 for the model = 59% for Tcr group and 44% for CsA group. Age, gender (male = 1; female = 2), plaque scores, months since transplant, and prednisolone, tacrolimus, and cyclosporin dosage were not included in the final model.

to a reduction in GO after conversion from CsA to Tcr.5,6 Nevertheless, GO has been reported in subjects medicated with Tcr,17 and this effect seems to be more severe in subjects also medicated with CCB.3,8 Our results revealed a 38.1% prevalence of GO in subjects under a CsA regimen, which is similar to previous reports,15,18,19 even though variations in

20.71

29.81

these values were reported.11,20,21 In relation to the Tcr group, our findings revealed a 17.9% prevalence of GO. Two studies showed no GO in subjects medicated only with Tcr,8 whereas another revealed a 15% prevalence.3 This variability in the prevalence of GO among subjects treated with CsA could be explained by the different methods applied to measure GO, the different parameters used to define its clinical significance, and the characteristics and number of individuals included in the study sample. It should be pointed out that the number of subjects examined in our study appears to be well founded compared to the total number of individuals in previous studies. In the present study, gingival overgrowth was observed in both groups, but the severity was greater in the CsA group. The mean percentage of GO score was 16.9% for the Tcr group and 29.03% for the CsA group. Similar findings have been reported for subjects under Tcr (14.1%) and CsA (22.4%) regimens.3 It should be emphasized that few authors define a cutoff point at which overgrowth is said to be present, nor do they define a level at which they regard the GO to be clinically significant,3,22,23 which might explain the wide range of prevalence results observed in the different studies.1,3,11,15,18-20 However, recent studies pointed to a 30% cutoff point for clinically significant GO.1,3 The results of the present study demonstrated that the mean percentage of GO score in those subjects who did not take CCB in the CsA and Tcr groups was 23.5% and 9.96%, respectively. Our findings also showed that the concomitant use of CCB increased 973


the prevalence of gingival enlargement in both groups. The increase of GO scores could be a consequence of the synergistic effect of the CCB.1,3,9,16 The fact that CCB alone induces GO could account for this raise in the scores of GO, and the pharmacokinetic interactions could also explain the higher values of GO found in subjects using CsA or Tcr in association with CCB. One study evaluating risk factors of GO in 100 subjects under a CsA regimen in the absence of CCB showed the presence of a 16.0% clinically significant GO.1 In the present study, the persistent GO in the Tcr group can most likely be explained by the concomitant use of CCB or the previous use of CsA. We verified that 62.5% of the Tcr subjects with clinically significant GO had a history of previous CsA use. However, we have no records of their gingival condition prior to Tcr medication. Further studies should be conducted to verify the periodontal status of the subjects before they undergo a Tcr regimen to elicit the true role of this drug in the GO. An increased risk of GO had been previously associated with plaque index and gingival inflammation.1,12,18 In our study, oral hygiene levels in subjects taking CsA or Tcr were similar, but the range of the papillary bleeding index was greater in the CsA group. Therefore, the differences were not statistically significant. Conflicting data on the association between time since transplant and GO has been reported.15,18,20 Time since transplant was not correlated to GO scores in this study for either CsA or Tcr regimens. It should be noted that the range in months for mean time since transplant was large for both groups. Although, a minimum 2-month time since transplant was considered, most cross-sectional studies have shown that GO occurs between 1 and 2 months since transplant.1,3,15,20,21 These questions should be better addressed by standardizing subjects with respect to the time since transplant in further studies. Some reports showed controversial findings regarding the correlation between serum immunosuppressive levels and GO. Our findings showed that serum drug levels were not correlated with GO, in accordance with other studies.3,4,16,18 In contrast, a positive correlation has been demonstrated.19-21 It should be emphasized that serum immunosuppressive levels vary notably during short periods of time, and longitudinal studies are also required to clarify this issue.22 Univariate analysis in the CsA group indicated that age, papillary bleeding index, plaque index, concomitant medication with CCB, and azathioprine dosage were significant variables associated with GO. In the Tcr group, the same risk variables were significant. Moreover, subjects with previous use of CsA showed more GO compared to those who had not taken the drug. When these risk variables were evaluated by re974


gression analysis in the CsA group, only azathioprine dosage, papillary bleeding index, and concomitant use of CCB remained significant (adjusted R2 = 44%). The same variables remained significant in the Tcr group, also including previous use of CsA in the final model (adjusted R2 = 59%). Some authors suggested that azathioprine has an anti-inflammatory property1,24 and that its concomitant use in immunosupressive regimens required significantly lower CsA dosages.1,21 The permanence of azathioprine in the multivariate final model as a risk factor of GO in the CsA and Tcr groups could be explained by the lower range dosage given to these groups of subjects. The positive relationship between GO and inflammation has been described;12,25 however, it is difficult to identify which side of the relation is causal in crosssectional studies. The presence of the papillary bleeding index and the absence of plaque scores in the final model further highlight the primary role of inflammation in this condition, as previously claimed.1 Therefore, it is reasonable to suggest that improved oral hygiene may at least minimize the severity of GO, reflecting the removal of the inflammatory component of this condition. Thus, an improved oral hygiene regimen failed to prevent the occurrence of CsA-induced GO in transplant subjects.26,27 The GO variability could not be explained by the plaque score because this variable failed to be maintained in the final multivariate model. The selection of predictor variables for a model is affected by the degree of correlation between the variables themselves, and highly correlated pairs are usually not both included unless they offer separate information.1,3,28 The removal of the plaque score in favor of the papillary bleeding index in these models is in line with earlier studies evaluating subjects medicated with Tcr and CsA.3,18 The concomitant use of CCB was a significant risk factor for the prevalence and severity of GO in the CsA and Tcr groups. CCB alone accounts for 30% of the variation in the GO score (R2 = 0.298) in subjects medicated with Tcr and for 27% in the CsA group (R2 = 0.271). Our sample appears to be soundly founded compared to the total number of individuals examined in previous studies. The findings from our study demonstrate that drug-induced GO is a significant problem in transplant subjects medicated with Tcr, although the severity is lower than that seen in subjects medicated with CsA. CONCLUSIONS The CsA group had a higher percentage of GO compared to the Tcr group, and the major risk factor for this condition seems to be concomitant medication


with CCB for both Tcr and CsA regimens. Other significant predisposing factors for the severity of GO are the papillary bleeding index and azathioprine dosage. Moreover, it was well established that previous medication with CsA was associated with GO in Tcr regimens. Tcr can also be a viable choice for those who show marked gingival enlargement under CsA regimens. As most studies investigate the GO related to either CsA or CCB, there appears to be a need for prospective clinical studies detailing the changes in the gingival health of subjects under a Tcr regimen in addition to potential associated risk variables.

REFERENCES 1. Thomason JM, Seymour RA, Ellis JS. Risk factors for gingival overgrowth in patients medicated with cyclosporin in the absence of calcium channel blockers. J Clin Periodontol 2005;32:273-279. 2. Miranda J, Brunet L, Roset P, Berini L, Farre´ M, Mendieta C. Prevalence and risk of gingival overgrowth in patients treated with diltiazen or verapamil. J Clin Periodontol 2005;32:294-298. 3. Ellis JS, Seymour RA, Taylor JJ, Thomason JM. Prevalence of gingival overgrowth in transplant patients immunosuppressed with tacrolimus. J Clin Periodontol 2004;31:126-131. 4. Seymour RA, Jacobs DJ. Cyclosporin and the gingival tissues. J Clin Periodontol 1992;19:1-11. 5. James JA, Boomer S, Maxwell AP, et al. Reduction in gingival overgrowth associated with conversion from cyclosporin A to tacrolimus. J Clin Periodontol 2000; 27:144-148. 6. Hernańdez G, Arriba L, Lucas M, de Andres A. Reduction of severe gingival overgrowth in a kidney transplant patient by replacing cyclosporin A with tacrolimus. J Periodontol 2000;71:1630-1636. 7. Feehally J, Walls J, Mistry N, et al. Does nifedipine ameliorate cyclosporin A nephrotoxicity? BMJ 1987; 295:310. 8. James JA, Jamal S, Hull PA, et al. Tacrolimus is not associated with gingival overgrowth in renal transplant patients. J Clin Periodontol 2001;28:848-852. 9. McKaig SJ, Kell D, Shaw L. Investigation of the effect of FK506 (tacrolimus) and cyclosporin on gingival overgrowth following paediatric liver transplantation. Int J Paediatr Dent 2002;12:398-403. 10. Thomason JM, Seymour RA, Rice N. The prevalence and severity of cyclosporin and nifedipine induced gingival overgrowth. J Clin Periodontol 1993;20: 37-40. 11. Vescovi P, Meleti M, Manfredi M, Merigo E, Pedrazzi G. Cyclosporin-induced gingival overgrowth: A clinicalepidemiological evaluation of 121 Italian renal transplant recipients. J Periodontol 2005;76:1259-1264. 12. Seymour RA, Ellis JS, Thomason JM. Risk factors for drug-induced gingival overgrowth. J Clin Periodontol 2000;27:217-223. ¨ e H. Periodontal disease in pregnancy: 13. Silness J, Lo Correlation between oral hygiene and periodontal condition. Acta Odontol Scand 1964;22:131-135.


14. Saxer UP, Mu ¨ hlemann HR. Motivation and education (in German). SSO Schweiz Monatsschr Zahnheilkd 1975; 85:905-919. 15. Pernu HE, Pernu LMH, Huttunen KRH, Nieminen PA, Knuuttila LE. Gingival overgrowth among renal transplant recipients related to immunosuppressive medication and possible local background factors. J Periodontol 1992;63:548-553. 16. Spratt H, Boomer S, Irwin CR, et al. Cyclosporin associated gingival overgrowth in renal transplant recipients. Oral Dis 1999;5:27-31. 17. Williams R, Neuhaus P, Bismuth H, et al. Two-year data from the European multicentre tacrolimus (FK506) liver study. Transpl Int 1996;9(Suppl. 1):S144-S150. 18. Thomason JM, Seymour RA, Ellis JS, et al. Determinants of gingival overgrowth severity in organ transplant patients. An examination of the role of HLA phenotype. J Clin Periodontol 1996;23:628-634. 19. Wilson RF, Morel A, Smith D, et al. Contribution of individual drugs to gingival overgrowth in adult and juvenile renal transplant patients treated with multiple therapy. J Clin Periodontol 1998;25:457-464. 20. Thomas DW, Newcombe RG, Osborne GR. Risk factors in the development of cyclosporine-induced gingival overgrowth. Transplantation 2000;69:522-526. 21. Somacarrera ML, Hernandez G, Acero J, Moskow BS. Factors related to the incidence and severity of cyclosporine-induced gingival overgrowth in transplant patients. A longitudinal study. J Periodontol 1994;65: 671-675. 22. Cebeci I, Kantari A, Gurel N, et al. Analysis of peripheral blood leukocytes in patients with cyclosporine A-induced gingival hyperplasia. J Periodontol 1998; 69:1435-1439. 23. Thomason JM, Seymour RA, Rawlins MD. Incidence and severity of phenytoin-induced gingival overgrowth in epileptic patients in general medical-practice. Community Dent Oral Epidemiol 1992;20:288-291. 24. Altman DG. Practical Statistics for Medical Research. London: Chapman & Hall; 1991:325-364. 25. Seymour RA, Thomason JM, Ellis JS. The pathogenesis of drug-induced gingival overgrowth. J Clin Periodontol 1996;23:165-175. 26. Seymour RA, Smith DG. The effect of a plaque control programme on the incidence and severity of cyclosporin-induced gingival changes. J Clin Periodontol 1991;18:107-110. 27. McGaw T, Lam S, Coates J. Cyclosporin-induced gingival overgrowth: Correlation with dental plaque scores, gingivitis scores, and cyclosporin levels in serum and saliva. Oral Surg Oral Med Oral Pathol 1987;64:293-297. 28. Seymour RA, Smith DG, Rogers SR. The comparative effects of azathioprine and cyclosporin on some gingival health parameters of renal transplant patients: A longitudinal study. J Clin Periodontol 1987;14:610-613. Correspondence: Dr. Fernando de Oliveira Costa, Department of Periodontology, Federal University of Minas Gerais, Antonio Carlos Ave., 6627 Pampulha, P.O. Box 359, 31270-901 Minas Gerais, Belo Horizonte, Brazil. Fax: 55-31-3282-6787; e-mail: [email protected]. Accepted for publication December 8, 2005.

975