There is a large body of evidence that patients with Balkan Endemic Nephropathy (BEN), in contrast to other chronic kidney diseases 12, do not manifest increased blood pressure 3456789. However, little is known about the development of blood pressure in BEN patients, in particular, whether increased blood pressure occurs in preclinical stages. BEN was first characterized in the Vratza District, Bulgaria in 1956 10. Later, similar nephropathies were described in Yugoslavia in 1957 310 and in Romania in 1961 11. As a result, in 1964, the disease was recognized as a new nosological entity and was named Balkan Endemic Nephropathy. BEN is a tubulo-interstitial kidney disease that progresses slowly for many years. The final disease stage is characterized by renal failure and shrinkage of both kidneys to 3–4 cm in length 12. Recent studies have shown that the offspring of BEN patients have smaller kidney lengths or cortex widths 1314, which may be a marker of a reduced number of nephrons 1516. Mackenzie et al. advanced the hypothesis that the total nephron number is a factor determining susceptibility to increased blood pressure 17. In addition, there is evidence that the number of nephrons co-determines kidney dimensions 18. Lastly, Singh et al. and Zumrutdal et al. have shown that decreased kidney dimensions are associated with increased systolic blood pressure 1619.

We conducted a prospective study in the Vratza District of Bulgaria, a rural area, and measured the kidney cortex width, kidney length, and blood pressure of the offspring of BEN patients and control offspring at two time points, one year apart. We hypothesize the following:

- A smaller kidney cortex width is related to a higher average systolic blood pressure and a higher pulse pressure.

- An offspring with parental history of BEN has a higher average systolic blood pressure and a higher pulse pressure.

Of the 201 participants in the first investigations (2003/04), 189 participated in the follow-up (2004/05, 94%). Two were deceased, one moved out of the area, two could not be contacted, and 7 decided not to participate further. Of the 12 lost, 9 were offspring of BEN patients. Potential risk factors show similar distributions in each of the two investigations (Table 1). One participant had newly developed diabetes. Comparison offspring were about 1.5 to 2 cm taller than BEN offspring; however, the latter were on average 1.5 kg heavier. Blood pressure seems to be higher in BEN offspring: 30.4% were classified as having moderate hypertension (systolic blood pressure ≥140 and/or diastolic ≥90 mm Hg) in the investigation in 2003/04 and 33.3% in 2004/05 (Table 1). The prevalence of moderate hypertension in our rural sample, with a mean age of 50 years, is only marginally lower than the proportion reported in another study of urban population of the same age in Bulgaria 25. Also, offspring of BEN patients reported more antihypertensive treatment in the year before the investigation.

Figure 1 shows the agreement between the two systolic blood pressure measurements one year apart. The intraclass correlation coefficient (ICC) of 0.68 (lower 5% limit: 0.61) was good for systolic blood pressure, but lower for diastolic blood pressure (ICC = 0.49, lower 5% limit: 0.39). For the minimal cortex width, the ICC was 0.66 (lower 5% limit: 0.58) and 0.92 for kidney length (lower limit: 0.89). Blood lead concentration also showed good agreement (ICC = 0.65, lower limit: 0.57).

The study population consists of 201 subjects that provided 388 observations (two measurements for most participants, one year apart). When analyzing blood pressure differences while controlling for confounders, cortex width and parental BEN status had significant effects on systolic and pulse pressure, but not on diastolic blood pressure. A decrease in the minimal kidney cortex of 1 mm was linearly related to an increase in systolic blood pressure of 1.4 mm Hg (p = 0.005) and an increase of pulse pressure of 1.1 mm Hg (p = 0.002).

Figure 2 shows the scatter plot of minimal cortex width and pulse pressure using a cubic regression with 95% confidence limits. A cubic regression fitted the observations best when no other predictors were taken into consideration. Within the boundaries of our measurements of the cortex width there was no decrease in blood pressure once the minimal kidney cortex width was 16 mm or larger. Blood pressure started to increase when the width was smaller than 16 mm.

A maternal history of BEN was associated with an increased systolic blood pressure in offspring of 6.7 mm Hg (p = 0.03), and when both parents had BEN, an increased systolic blood pressure of 9.9 mm Hg (p = 0.002). There was no statistically significant effect of a paternal history of BEN.

To investigate whether the use of antihypertensive medication had distorted our findings, we repeated the model excluding those observations with antihypertensive drug use. This reduced the number of repeated observation from 388 to 322 (176 subjects), but did not change our findings.

Our findings show that a decrease of the minimal kidney cortex width of 1 mm was related to an increase in systolic blood pressure of 1.4 mm Hg. There was no association between kidney length and blood pressure. A maternal history of BEN was associated with a significant increase in systolic blood pressure of 6.7 mm Hg, paternal BEN with a non-significant increase of 3.2 mm Hg, and both parents affected, a significant increase of 9.9 mm Hg. The lack of association of either kidney cortex width or a parental history of BEN with diastolic blood pressure may either stem from the larger individual variance of this parameter or may indicate that there is no such link.

In a study with healthy volunteers in England, Raman et al. did not identify a clear pattern between measurements of the right and left kidneys and blood pressure 26. In another population-based study, Paivansalo et al. found that the kidneys of hypertensive women were slightly larger (p < 0.08) 27. To the contrary, comparing hypertensive and normotensive Turkish patients, Zumrutdal et al. found that kidney length and volume were significantly smaller in hypertensive patients 19. Thus, it is not clear whether hypertension affects kidney dimensions or vice versa.

For adult Aboriginals in Australia, Singh et al. found that a 10 mL decrease in kidney volume was associated with a 0.5 mm Hg increase in systolic blood pressure 16. The study of Singh et al. did not provide information on cortex width; in our study we do not have information on kidney volume. To compare results, we expressed both findings as a percent decrease in kidney dimensions. A 10 mL decrease in kidney volume in the Australian study represents approximately a 5% decrease in volume (range: 62 – 267 mL). In our study of adult Caucasian offspring from Bulgaria, a 5% decrease in the minimal cortex width (0.5 mm, range 10 – 20 mm) is associated with a 0.7 mm Hg increase in blood pressure. Thus, both studies found a similar effect of kidney dimensions on blood pressure. Singh et al. also showed a cubic-type association between kidney volume and systolic blood pressure. Similarly, we have demonstrated a cubic regression between minimal kidney cortex width and blood pressure; the threshold we detected in our model, is approximately 16 mm of cortex width. However, if we control for other variables in mixed regression models, a straight line is also appropriate (Table 2).

Mackenzie et al. posited the hypothesis that the total nephron supply at birth is a factor in determining a susceptibility to increased blood pressure 17. Given that kidney dimensions correlate with the number of nephrons 18, our finding that minimal cortex width is significantly associated with systolic and pulse pressure supports Mackenzie's hypothesis. In addition, we found that, in offspring, maternal BEN is associated with increased systolic and pulse pressure. In a prior report of this project, we showed that a maternal but not a paternal history of BEN is related to reduced cortex width in their adult offspring 14. The stronger role of maternal BEN in offspring susceptibilities (cortex width and average blood pressure) and the lack of paternal-related risks may indicate, in our opinion, the importance of feto-maternal interactions during pregnancy. Mothers who later develop BEN do not yet have the disease during their childbearing years. However, they are likely to have developed precursors of the disease before becoming pregnant, which, however, must be sufficient in themselves to initiate a higher risk in the offspring for having a smaller kidney cortex width and increased systolic and pulse pressure.

Hence, a maternal history of BEN may have both a direct and an indirect effect on blood pressure in offspring. Indirectly, maternal BEN seems to decrease kidney cortex width, which in turn increases blood pressure in adult offspring. Directly, maternal BEN seems to initiate regulations likely to raise the average blood pressure in affected offspring. However, hypertension is not a predominant feature in BEN offspring. Thus, it is unlikely that the reduction in kidney cortex width in BEN offspring results from hypertension.

Renal mass is considered to contribute directly to the development and maintenance of hypertension 161928. Also renal scarring, an index of chronic damage, has also been related to blood pressure 29. Since we have previously reported a reduced kidney cortex width in BEN offspring 14, our finding that minimal cortex width is related to blood pressure is in agreement with prior reports.

However, in BEN patients, whose kidneys shrink, the lack of increased blood pressure has been considered to be a typical characteristic before the disease reaches its final stage 3456789. The reason for this unusual association has never been explained. We were surprised to find that BEN offspring had an average increased systolic blood pressure, in particular related to a maternal history of BEN. In agreement with our finding, Arsenovic et al. recently also reported an increased prevalence of hypertension (20 out of 47 persons) in BEN family members 30.

Since our data is based on adult offspring, we are uncertain about the future development of blood pressure in this group. There are three possibilities: (I) Offspring of BEN patients will not develop BEN but maternal BEN may lead to subclinical changes such as an increase in average blood pressure. An argument against this option is that we have found reduced kidney sizes and increased protein excretion in the BEN offspring, which may be considered as early signs of BEN 14. (II) BEN offspring will develop BEN and their higher average blood pressure will disappear with the progression of BEN. One speculation is that the juxtaglomerular cells become injured and thus the renin-angiotensin system also is impared, which may prevent hypertension. (III) A third option is that the reported lack of hypertension in BEN patients was not based on scientifically sound conclusions and needs to be revised.

We found that adult offspring of BEN patients have an average increased blood systolic pressure and pulse pressure, determined in two repeated measurement one year apart. This is contrary to reports in BEN patients, who do not have this characteristic. Surprisingly, the increase was related to a maternal history of BEN, or a history of BEN in both parents, but not to a paternal history alone. Increased blood pressure was also associated with a reduction of the minimal width of the kidney cortex, but not with a reduction in kidney length. In our understanding, these findings emphasize the role of maternal priming in determining susceptibility for increased blood pressure. Balkan endemic nephropathy may serve as a model to better understand the development of both kidney disorders and their related blood pressures.

The author(s) declare that they have no competing interests.

PSD developed and designed the study and helped write the report. SDT assisted in analyzing the data, literature review, and writing the discussion. RBG conducted the analyses of lead in blood, and assisted in writing the discussion. VAS conducted the clinical examinations of the study subjects and contributed to the interpretation. AGB conducted the ultrasound examinations of the study subjects and contributed to the interpretation. WJK helped design the study and conducted the statistical analysis and worked on the manuscript. All authors read and approved the final manuscript.