|Year : 2014 | Volume
| Issue : 2 | Page : 343-352
|Incidence and predictive factors of Balkan Endemic Nephropathy: A longitudinal study
Kesinee Hanjangsit1, Plamen S Dimitrov2, Hongmei Zhang1, Vecihi Batuman3, Jim Burch1, Svetla D Tzolova2, Wilfried J Karmaus1
1 Department of Epidemiology and Biostatistics, University of South Carolina, Columbia, South Carolina, USA
2 National Center of Public Health Protection, Sofia, Bulgaria
3 Section of Nephrology-Hypertension, Tulane University Medical Center, and VA Medical Center, New Orleans, USA
Click here for correspondence address and email
|Date of Web Publication||11-Mar-2014|
| Abstract|| |
Balkan endemic nephropathy (BEN) is a chronic kidney disease that progresses slowly. There are no known clinical markers to identify an early disease development. We evaluated the relationship between parental history of BEN and clinical markers as predictors of new occurrences of BEN. A 5-year prospective study in the offsprings of BEN and control patients was conducted in Vratza, Bulgaria, between 2003 and 2009 using markers in years one and three to predict new cases of BEN in the year five. We defined incident cases of BEN based on parental history, reduced kidney size and reduced kidney function, distinguishing probable and definite BEN, both combined as total incidence. The data were analyzed by Cox regression models using age as time scale and controlling for gender. We estimated hazard ratios and their 95% confidence intervals. The incidence of BEN was 17.4%. Paternal history was strongly associated with all three incidence groups (hazards ratio: 27-68, P <0.05). A reduction of kidney size of 1 mm resulted in a 5% increased hazard. However, taking parental history of BEN into account, these associations lost their significance. No kidney function measures were associated with new onset of BEN. A parental history of BEN is more important than clinical markers predicting the incidence of BEN. Without this information, kidney length forecasts probable BEN and the total incidence, while none of any clinical markers was related to definite BEN.
|How to cite this article:|
Hanjangsit K, Dimitrov PS, Zhang H, Batuman V, Burch J, Tzolova SD, Karmaus WJ. Incidence and predictive factors of Balkan Endemic Nephropathy: A longitudinal study. Saudi J Kidney Dis Transpl 2014;25:343-52
|How to cite this URL:|
Hanjangsit K, Dimitrov PS, Zhang H, Batuman V, Burch J, Tzolova SD, Karmaus WJ. Incidence and predictive factors of Balkan Endemic Nephropathy: A longitudinal study. Saudi J Kidney Dis Transpl [serial online] 2014 [cited 2019 Nov 18];25:343-52. Available from: http://www.sjkdt.org/text.asp?2014/25/2/343/128539
| Introduction|| |
Balkan endemic nephropathy (BEN) is a chronic tubulointestitial disease affecting populations in several areas in Bulgaria, Bosnia-Herzegovina, Croatia, Romania and Serbia. ,,, The disease has been described in adults and cases predominantly occur after the age of 50 years. Some studies found that the prevalence of the disease among men and women is equal;  however, other studies showed that women are slightly more likely than men to develop the disease. , The etiology of BEN is unknown. BEN is relatively asymptomatic, with a long latency period, and no diagnostic test is presently available to identify those at an increased risk or for early detection of BEN. The clinical diagnosis is based on excluding other renal diseases, certain epidemiological features and a set of various clinical criteria. ,, Early clinical markers are increased excretion of low molecular weight proteins, especially of β 2 -microglobulin in urine, reduced creatinine clearance and symmetrically decreased kidney measures. , However, no study has yet identified clinical markers that are able to predict the disease in a prospective study.
β 2 -microglobulin is filtered through the glomerular capillary wall and completely reabsorbed and metabolized by proximal tubules. , Because β 2 -microglobulin is a major constituent of tubular proteinuria, an elevated concentration of β 2 -microglobulinuria is considered an indicator of proximal renal tubular dysfunction.  Increased excretion of β 2 -microglobulin has been found in BEN patients ,, and in their healthy offsprings, , especially in those whose mother had BEN.  One study concluded that β 2 -microglobulinuria can characterize kidney malfunction but that it was not specific for BEN.  However, the consistent finding of β 2 -microglobulinuria in BEN patients led to the speculation that it may be more of a marker of current disease. 
Regarding renal function, Djukanovic et al  showed that in comparison with other kidney diseases, glomerular filtration rate was higher in BEN patients in the early stage while another study claimed that creatinine clearance was normal.  Mitic-Zlatkovic et al  reported a reduction in creatinine excretion among BEN patients.
A small kidney size is one of the important characteristics of BEN, especially in advanced stages. Lezaik et al concluded that compared with other kidney diseases, BEN patients had smaller kidneys.  However, it is unclear when the shrinkage starts. Some studies suggest that kidney sizes are normal in patients without renal failure, but decrease remarkably as the disease progresses. , Recent studies, conversely, have shown that a reduction in kidney length appears in the early stage. , Our previous results found a reduction of kidney length in healthy offspring with a family history of BEN. , In addition, smaller kidney cortex widths were identified among participants with a family history of BEN prior to disease onset. , Maternal history of BEN was a stronger predictor of kidney size than paternal history. ,
BEN is related to family history, in which several members of one family or several generations may be affected, although no specific genetic marker has been identified. , A case-control study showed that individuals whose mother had BEN were 48-times more likely to have the disease than other BEN patients, which suggests that a maternal history is an important risk factor, and the possible existence of a sex-linked trait.  A person whose father had BEN was about 8.8-times more likely to develop the disease. This finding motivated us to evaluate whether maternal history of BEN was a stronger predictor of BEN than paternal history.
The goal of this study aim was to understand the clinical pathogenesis that leads to new cases of BEN using a longitudinal design. In particular, the aim was to evaluate the influence of clinical markers such as β 2 -microglobulin, creatinine clearance, kidney length and cortex width to predict the occurrence of BEN. We chose to study a cohort of adult offspring of BEN patients as this group is at high risk, and we expected a substantial incidence during the five-year observation period. The offspring control group had parents with other diseases.
| Subjects and Methods|| |
Study design and population
Our study was conducted in Bulgaria and is described in detail elsewhere. , Briefly, in 2001, 104 adult offspring with a parental history of BEN were recruited from a hospital database in Vratsa, Bulgaria, an endemic region. The parental history of BEN was based on epidemiology, clinical or laboratory and pathological/anatomical criteria.  Adult offspring of non-BEN parents (97 study subjects) were frequency-matched with regard to gender and ten-year age groups, and were recruited into the study as the control group. Both the study groups were enrolled and examined for the first time in 2003-2004 and re-examined four times between 2004 and 2009. In the third examination (2005-06), 18 additional participants of BEN patients were recruited. During the same examination year, a parent of one control participant developed BEN; therefore, this participant was moved into the BEN offspring group for analysis of this and subsequent years.
All participants provided written consent. The institutional review board of the National Center of Public Health Protection, Sofia, Bulgaria and of the University of South Carolina approved the study.
We conducted face-to-face interviews with all participants either in the hospital or by visiting their home village. A standardized questionnaire included family history of BEN and, to calculate age at each examination, the date of birth. Hypertension was defined as a systolic blood pressure 140 mm Hg and/or a diastolic blood pressure 90 mm Hg and/or history of hypertension. To identify diabetes, both history of diabetes and serum glucose were used. If the participants reported that they had diabetes within five years preceding the study period, they were defined as diabetic. If the serum glucose was recorded to be >6.2 mmol/L during the previous two years, those participants were also classified as diabetic. Collection of urine sample for examination of proteinuria has been described elsewhere.  Briefly, urine specimens were taken during a 4-h examination. p 2 -microglobulin was measured in aliquots of the hour-one urine sample while total protein and albumin were measured in the hour-four urine sample. Kidney sizes were measured in millimeters using ultrasound investigations of both kidneys (left and right). The longest dimension and the smallest thickness of kidney parenchyma were determined. Also, location, size and morphology (cysts, stones and tumors) of the kidneys were obtained. All images were saved electronically for future reference. Serum creatinine was collected and was used to calculate creatinine clearance (CCR) using the Cockcroft and Gault formula (1976). 
Definition of BEN incidence
Incident BEN cases were defined using the following definitions during the year 2003-2004: (1) parental history of BEN, (2) reduced kidney size (kidney length <100 mm, kidney cortex width <11 mm) and (3) reduced kidney functions (serum creatinine >124 μmol/L for males and > 106 μmol/L for females; creatinine clearance using the Cockcroft and Gault formula <60 mL/min; β 2 -microglobulin >0.2 mg/g creatinine). Definite cases of BEN required a parental history of BEN, a reduced kidney length and/or cortex width and two or more of the persisting indicators of reduced kidney functions listed above. Probable BEN cases were defined as those with a parental history of BEN, a reduction in either kidney length or cortex width and a reduction in at least one kidney function marker in repeated measurements. These criteria represent an operational definition for epidemiological studies. ,,
| Statistical Analysis|| |
Cox proportional hazards was used to estimate crude and adjusted hazard ratios and their 95% confidence intervals (CI) for the association between clinical markers or parental history of BEN and BEN incidence after controlling for gender. We estimated the effects of parental history of BEN and clinical markers in the same model. This approach assumes that none of the clinical markers acted as intervening variables, but that their contribution to the prediction of BEN could be assessed independently. Penalized maximum likelihood estimation was applied in the case of the monotone likelihood.  Participants who developed either definite or probable BEN were defined as a total incidence group. The association between the main predictors of interest and three outcomes (total incidence, definite or probable BEN) was estimated in separate statistical models. We used age at disease development as a time-to-event variable in the analysis. Offspring without disease at age of their last follow-up (2008-2009) or participants who dropped out were censored. Therefore, the time-to-event (age at onset) is absorbed into the unspecified baseline hazard. ,
The role of clinical markers including kidney length and cortex width, CCR and p 2 -microglobulin were determined separately for two periods, 2003-2004 and 2005-2006, in the survival analysis. All clinical markers had a continuous scale. Because of a skewed distribution, the natural logarithm of urinary p 2 -microglobulin, urinary total protein and urinary albumin concentration was used. Interactions between gender or parental history of BEN with each clinical marker were also assessed. A two-sided P 0.05 was considered statistically significant. Model fits and outliers were identified using residual analysis. All covariates were initially assessed for time-dependent effects using a test based on the Schoenfeld residuals, a standard score process test and time-dependent interaction. , The proportional hazard assumption was not violated in any model. Intraclass correlation coefficients (ICCs) were used to examine relationships between continuous clinical measurements. Analyses were conducted in SAS 9.2. 
| Results|| |
Of the 201 participants enrolled after the first investigation (2003-2004), 94% remained in the second investigation (2004-2005). In 2005-2006, 18 participants (offspring of BEN patients) were newly recruited whereas seven participants dropped out, resulting in 200 participants in the third round of data collection. Seven participants withdrew during the fourth examination period. One hundred and seventy-seven of 219 (80.8%) participants remained in the study up to the last year (data not shown).
Over the five-year study period, 38 (17.4%) offspring developed BEN, while 6.8% and 10.5% developed definite and probable BEN, respectively. The characteristics of the participants in 2005-06 are shown in [Table 1]. There were no significant differences in the distribution of age, gender, diabetes, hypertension, urine total protein, urine albumin and urine β 2 -microglobulin among offspring of BEN and offspring of non-BEN (P >0.05) [Table 1]. Offspring with parental history of BEN had smaller kidney length (P <0.004) and cortex width (P = 0.028) than those of offspring of non-BEN [Table 1]. In addition, all new cases of BEN have a positive parental history [Table 1]. Maternal history of BEN was more frequently observed among definite BEN cases (53.3%), while paternal history of BEN was more common among probable BEN cases (39.1%). Participants with a definite BEN diagnosis had the oldest age, the smallest kidney length and cortex width, the lowest CCR and the highest β 2 -microglobulin. There was no gender difference among definite BEN cases, but women were more likely than men to develop probable BEN. All BEN offspring resided in endemic areas, most of them more than 20 years (data not shown).
|Table 1: Characteristic of participants stratified by parental history of BEN in the year 2005–06.|
Click here to view
Almost all clinical markers were moderately and statistically significantly correlated. Only maternal BEN correlated with kidney length (r = -0.176, P = 0.009). All markers and the parental history of BEN were significantly correlated with age (data not shown). ICC shows a high agreement between measurements of kidney length and cortex width in the left and right kidneys (0.99) and over the course of two repeated assessments (0.98, data not shown). Regarding kidney function (CCR and β 2 -microglobulin), the ICCs over the course of two repeated examinations were 0.99 for both markers (data not shown).
The Kaplan-Meier product-limit method [Figure 1], which is used for estimating the cumulative functions, indicates a median age of 70 years to develop BEN (the sum of definite and probable BEN) for all participants (95% CI, 67-72). The curve shows that increased incidence started at 45 years, slowly declined between 50 and 60 years and steeply declined after 60 years of age. The range when BEN offspring developed BEN is between 45 and 78 years of age [Figure 1]a. Stratified by gender in [Figure 1]b, the median age until the onset of BEN in men tends to be higher than those of women (70 vs. 65 years old); however, the difference is not significant (P = 0.056). Among those with a parental history of BEN, the median age at diagnosis of BEN for participants whose father, mother or both had BEN was 61, 67 and 66 years, respectively [Figure 1]c.
[Table 2] provides the crude associations between covariates and first occurrence of BEN. Compared with the non-parental BEN control group, those with paternal BEN had the strongest association with all three outcomes evaluated (P <0.01). A decrease of 1 mm of kidney length increased the hazard ratio for new cases by about 6-7% for all three BEN outcomes, and a small kidney cortex width was associated with definite and incident BEN (P <0.05), whereas CCR, β 2 -microglobulin, hypertension, diabetes or gender were not related to any of the BEN outcomes (probable, definite or total incidence). Total urine protein and urine albumin were associated with probable and total incidence of BEN (P < 0.05).
The effect of parental history of this disease, adjusted for gender, hypertension, diabetes and clinical markers, is provided in [Table 3]. Compared with the non-parental BEN control group, the risk for definite BEN was about 27-times higher for those with a paternal history of BEN and the risk for first occurrence of probable BEN was about 62-times higher among those with a paternal history of BEN. The risk of developing definite BEN was about 26-times and of developing probable BEN was about 23-times higher among those with a maternal history of BEN compared with those without a history. Clinical markers, hypertension, diabetes and gender were not associated with any type of BEN (P >0.05) [Table 3].
|Table 3: Associations between clinical markers and definite, probable or incident BEN after adjustment for gender and parental BEN.|
Click here to view
We hypothesized that clinical markers may be intervening variables between a family history of BEN and BEN development; therefore, we Re-analyzed the model without parental history of BEN. We found that kidney length was associated with probable BEN and with total BEN incidence (P = 0.011 and 0.003, respectively). Kidney function measures, kidney cortex width and proteinuria were not associated with BEN incidence after controlling for gender, hypertension and diabetes (data not shown). To test whether clinical markers predicted BEN among those with a parental history of BEN, we repeated the analyses excluding those without a parental history of BEN. This approach reduced the number of observations from 219 to 104, but did not change our finding that no clinical markers were associated with any of the three BEN outcomes after adjustment for the effects of parental history. In addition, maternal or paternal history of BEN did not differ in their ability to predict BEN in offspring compared with biparental BEN (data not shown).
| Discussion|| |
The incidence of BEN over a five-year period in a cohort of BEN offspring among parent(s) with BEN aged 35-78 years of age was 17.4%. Survival analyses showed that the incidence started at 45 years. The proportion of disease-free offspring then slowly declined between 50 and 60 years of age, and steeply declined after 61 years of age. A paternal history of BEN was important with regard to the onset of BEN (hazard ratio = 32), while participants with a maternal history were 13-times more likely to develop BEN. Ignoring the parental history, our data indicate that kidney size was strongly associated with BEN incidence. In particular, kidney length was related to an increased risk of probable BEN. However, these clinical markers lost their importance once the parental history was taken into consideration, indicating that the clinical markers act as intervening variables between parental history and incidence of BEN in offspring.
The findings are unlikely to be explained by a selection bias as the follow-up proportion was high (80.8%). Regarding potential information biases, reliability among the kidney measurements indicated good agreement. In addition, a combination of several different criteria was used to define the BEN outcomes; hence, it is unlikely that the definition of BEN is affected by information biases.
One of the main limitations of this investigation is that other kidney dimensions were not measured, which would have allowed for the calculation of kidney volume. Although kidney volume is the most accurate method for measuring kidney size,  it is not widely used because it is difficult to measure.  Some studies noted that kidney length is a good indicator of kidney size and it provides adequate information for routine clinical examinations. , Djukanovic et al  showed that the kidney volume was not superior compared with kidney length in distinguishing BEN from healthy controls or patients with other kidney disease. Dimensions of the kidney cortex, however, are considered by some to be more sensitive markers for atherosclerotic kidney diseases. 
We found, in accordance with other studies, , that parental history was the most important predictor of an increased risk for BEN diagnosis in this prospective study. Unexpectedly, paternal history of BEN was more strongly associated with the onset BEN than maternal history, which is not consistent with our case-control study conducted in 2001  nor with our reports of a strong association between clinical markers and BEN in offspring with a maternal history of BEN. ,, Unfortunately, other studies have not distinguished paternal and maternal history. Thus, our results cannot easily be compared with other reports. Further studies are needed to investigate this difference.
It has been conclusively demonstrated that a reduction of kidney length is associated with an increased risk of BEN. ,,, Results from the present study are consistent with these observations while taking the appropriate temporal sequence into account. The prediction of BEN by kidney length measured at least three years before BEN manifestation supports the notion that a reduction of kidney length appears in the early stage of BEN. ,, However, the association between both kidney structure and occurrence of BEN lost their significance after adjusting for parental history of BEN. The strongly elevated risk among those with a parental history of BEN suggests that clinical predictors are less likely to improve the detection rates for BEN. However, it remains unclear how family history influences the development of this disease. Our results suggest that it is likely that changes in the kidney morphology represent an intervening step between family-related susceptibilities and the new development of BEN. Additional studies are needed to further examine these relationships.
Neither of the two kidney function tests analyzed in this study predicted BEN. Stefanovic et al  observed an increase in β 2 -microglobulin in both patients and healthy offspring, , and suggested that β 2 -microglobulin is an indicator of renal-tubular dysfunction, but not adequate to verify the diagnosis of BEN. Although CCR was not statistically significantly associated with the incidence of BEN, Mitic-Zlatkovic et al have shown that CCR is declined among BEN patients.  However, another cross-sectional study reported that creatinine clearance was elevated among BEN patients.  Because cross-sectional studies cannot evaluate the temporal sequence,  it is possible that abnormal kidney function may manifest later in the development of BEN. This may imply that morphological changes precede alterations of kidney function. To test this hypothesis, evaluations of the temporal relationship between kidney sizes and function in a longitudinal study are needed.
In conclusion, our results show that no clinical markers are needed to predict the occurrence of BEN if the parental status of BEN is known (hazards ratio = 49, 37 and 62 for biparental, maternal and paternal BEN, respectively). Without information on parental history of BEN, kidney length was the best predictor of probable BEN and the sum of definite and probable BEN (total incidence). None of the clinical markers was the best predictor for definite BEN.
| Acknowledgment|| |
The research was supported by NIH Research Grant # R01 TW06192 funded by the Fogarty International Center and the National Institute of Environmental Health Sciences, National Institutes of Health, USA, awarded to the National Center of Public Health Protection, Sofia, Bulgaria.
| References|| |
|1.||Dimitrov PS, Simeonov VA, Stein AD. Balkan endemic nephropathy in Vratza, Bulgaria, 1964-1987: An epidemiologic analysis of population-based disease registers. Eur J Epidemiol 2001; 17:847-53. |
|2.||Marie I, Bukvic D, Bogdanovic M, et al. Cross-sectional study in the Balkan endemic nephropathy village of Vreoci (Serbia). BANTAO J 2006;4:5-8. |
|3.||Ceovic S, Plestina R, Miletic-Medved M, Stavljenic A, Mitar J, Vukelic M. Epidemiological aspects of Balkan endemic nephropathy in a typical focus in Yugoslavia. IARC Sci Publ 1991:5-10. |
|4.||Cvoriscec D, Ceovic S, Borso G, Rukavina AS. Endemic nephropathy in Croatia. Clin Chem Lab Med 1998;36:271-7. |
|5.||Batuman V. Fifty years of Balkan endemic nephropathy: Daunting questions, elusive answers. Kidney Int 2006;69:644-6. |
|6.||Tatu CA, Orem WH, Finkelman RB, Feder GL. The etiology of Balkan endemic nephropathy: Still more questions than answers. Environ Health Perspect 1998;106:689-70. |
|7.||Schiller A, Gusbeth-Tatomir P, Pavlovic N, Ferluga D, Spasovski G, Covic A. Balkan endemic nephropathy: A still unsolved puzzle. J Nephrol 2008;21:673-80. |
|8.||Stefanovic V, Cosyns J. Balkan nephropathy. Oxford: Oxford University Press; 2004. |
|9.||Stefanovic V, Jelakovic B, Cukuranovic R, et al. Diagnostic criteria for Balkan endemic nephropathy: Proposal by an international panel. Ren Fail 2007;29:867-80. |
|10.||Danilovic V. Dijagnoza endemske nefropatije. Rodovi 62-XLVIII Od med nauka Akad nauka i umj BiH, Sarajevo 1973;17:53-64. |
|11.||Berggard I, Beam AG. Isolation and properties of a low molecular weight beta2-microglobulin occuring in human biological fluids. J Biol Chem 1968;243:4095-103. |
|12.||Karlsson F, Groth T SK. Turnover of human beta 2-microglobulin, the constant chain of HLA antigens. Eur J Clin Invest 1980;10:293-300. |
|13.||Schadurin G, Statius LW. Beta2-microglobulin: Its significance in the evaluation of renal function fluids. Kidney Int 1987;32:635-41. |
|14.||Hrabar A, Aleraj B, Ceovi S, Cvoriscec D, Vacca C, Hall PW 3rd. Beta 2-microglobulin studies in endemic Balkan nephropathy. Kidney Int Suppl 1991;34:S38-40. |
|15.||Radonic M, Radosevic Z. Clinical features of Balkan endemic nephropathy. Food Chem Toxicol 1992;30:189-92. |
|16.||Stefanovic V, Mitic-Zlatkovic M, Cukuranovic R, Mi|jkovic P, Pavlovic N, Vlahovic P. β2-microglobulin in patients with Balkan nephropathy and in healthy members of their families. Kidney Int Suppl 1991;40:S21-6. |
|17.||Dimitrov P, Tsolova S, Georgieva R, et al. Clinical markers in adult offspring of families with and without Balkan Endemic Nephropathy. Kidney Int 2006;69:723-9. |
|18.||Stefanovic V, Cukuranovic R, Djordjevic V, et al. Tubular marker excretion in children from families with Balkan nephropathy. Pediatr Nephrol 2009;24:2155-66. |
|19.||Batuman V. Possible pathogenetic role of low-molecular-weight proteins in Balkan nephropathy. Kidney Int Suppl 1991;34:S89-92. |
|20.||Djukanovic L, Bukvic D, Marie I. Creatinine clearance and kidney size in Balkan endemic nephropathy patients. Clin Nephrol 2004;61: 384-6. |
|21.||Trnacevic S, Halilbasic A, Ferluga D, et al. Renal function protein excretion and pathology of Balkan endemic nephropathy I Renal function. Kidney Int Suppl 1991;34:S49-51. |
|22.||Mitic-Zlatkovic M, Cukuranovic R, Lecic N, Stefanovic V. Urinary creatinine excretion in children from families with Balkan endemic nephropathy: Evidence for genetic predisposition to the disease. Pathol Biol 2000;48:554-7. |
|23.||Lezai V, Mari I, Jovanovi D, et al. Comparison of kidney size between patients with Balkan endemic nephropathy and other kidney di-seases. Kidney Blood Press Res 2008;31:307-12. |
|24.||Cukuranovic R, Savic V, Stefanovic N, Stefanovic V. Progression of kidney damage in Balkan endemic nephropathy: A 15-year follow-up of patients with kidney biopsy examination. Ren fail 2005;27:701-6. |
|25.||Stefanovic V, Polenakovic MH. Balkan nephropathy. Kidney disease beyond the Balkans? Am J Nephrol 1991;11:1-11. |
|26.||Ferluga D, Hvala A, Trnacevic S, et al. Pathology of Balkan Endemic Nepropathy-A Correlation with Established Kideny Disease Entities. Facta Univ Ser Med Biol 2002;9:82-7. |
|27.||Hanjangsit K, Dimitrov P, Karmaus W, et al. Reduced kidney size in adult offspring of Balkan endemic nephropathy patients and controls: A prospective study. Am J Med Sci 2010;340:94-102. |
|28.||Toncheva D, Dimitrov T, Stojanova S. Etiology of Balkan endemic nephropathy: A multifactorial disease? Eur J Epidemiol 1998;14:389-94. |
|29.||Karmaus W, Dimitrov P. Recent studies on Balkan Endemic Nephropathy in Bulgaria. Bulg J Public Health 2009;1:31-41. |
|30.||Karmaus W, Dimitrov P, Simeonov V, Tsolova S, Batuman V. Offspring of parents with Balkan Endemic Nephropathy have higher C-reactive protein levels suggestive of inflammatory processes: A longitudinal study. BMC Nephrol 2009;10:10. |
|31.||Cockcroft DW, Gault MH. Prediction of creatinine clearance from serum creatinine. Nephron 1976;16:31-41. |
|32.||Heinze G, Schemper M. A solution to the problem of separation in logistic regression. Stat Med 2002;21:2409-19. |
|33.||Kom EL, Graubard BI, Midthune D. Time-to-event analysis of longitudinal follow-up of a survey: Choice of the time-scale. Am J Epidemiol 1997;145:72-80. |
|34.||Hakulinen T, Tenkanen L. Regression analysis of relative survival rates. J R Stat Soc Ser C Appl Stat 1987;36:309-17. |
|35.||Therneau MT, Grambsch MP. Modeling Survival Data: Extending the Cox Model. New York: Springer-Verlag; 2000. |
|36.||Hosmer WD, Lemeshow S, May S. Applied Survival anlaysis: Regression Modeling of Time-to Event Data. Hoboken, New Jersey: John Wiley & Sons, Inc.; 2008. |
|37.||SAS-Institute. SAS/STATA Software. In. SAS Institute Inc, Cary, NC: 2009. |
|38.||Emamian SA, Nielsen MB, Pedersen JF, Rytte L. kidney dimensions at sonography: Correlation with age, sex, and habitus in 665 adult volunteers. AJR Am J Roentgenol 1993;160:83-6. |
|39.||Kang KY, Lee YJ, Park SC, et al. A comparative study of methods of estimating kidney length in kidney transplantation donors. Nephrol Dial Transplant 2007;22:2322-7. |
|40.||Griffiths G, Cartwright G, McLachlan M. Estimation of renal size from radiographs: Is the effort worthwhile? Clin Radiol 1975;26:249-56. |
|41.||Djukanovic L, Marie I, Marinkovic J, Ignjatovic S, Bukvic D. Evaluation of criteria for the diagnosis of Balkan endemic nephropathy. Ren Fail 2007;29:607-14. |
|42.||Mounier-Vehier C, Lions C, Devos P, et al. Cortical thickness: An early morphological marker of atherosclerotic renal disease. Kidney Int 2002;61:591-8. |
|43.||Toncheva D, Dimitrov T. Genetic Predisposition to Balkan Endemic Nephropathy. Nephron 1996; 72:564-9. |
|44.||Marinkovic D, Cvjeticanin S. Population-genetic study of Balkan endemic nephropathy in Serbia. Genetika 2007;43:1134-8. |
|45.||Gordis L. Epidemiology. 3 rd ed. Philadelphia, Pennsylvania: Elsevier Saunders; 2004. |
Wilfried J Karmaus
Department of Epidemiology and Biostatistics, University of South Carolina, Columbia, South Carolina, 29208
[Table 1], [Table 2], [Table 3]
| Article Access Statistics|
| Viewed||1309 |
| Printed||23 |
| Emailed||0 |
| PDF Downloaded||303 |
| Comments ||[Add] |