| Abstract|| |
Glomerulonephritis (GN) is a major cause of CRF in Nigerians. Experimental evidence and clinical studies mostly in Caucasian subjects have associated hydrocarbon (HC) exposure with GN. We conducted a case-control study using a questionnaire-based quantitative HC exposure measurement to compare lifetime HC exposure levels between Nigerian patients with GN-induced CRF and matched healthy control subjects. Fifty consecutive patients with CRF from GN were compared with age and sex matched group of 45 healthy controls. A questionnaire designed to assess the sources, duration and intensity of HC exposure was used to compute an HC exposure score (HES) for each participant and the scores for the two groups were then compared. The HES was significantly higher in the patients (score ± SEM) of 2307.5 ± 698.8 vs. 53.4 ± 16.5; p < 0.001. The HES was dichotomised by classifying all study subjects within the upper third of scores as a high-exposure sub-group. A significantly higher proportion of patients had high exposure (p<0.002). Logistic regression analysis excluded age and gender as confounding factors and determined a greater than four-fold risk of GN-induced CRF with high HC exposure (OR 4.3; 95% CI 1.7 - 11). In conclusion, our findings suggest that HC exposure is a significant risk factor for GN in Nigerians with CRF. Exposure limitation could help to reduce the burden of CRF in the country.
Keywords: Glomerulonephritis, CRF, hydrocarbon, Nigerians, exposure, risk assessment.
|How to cite this article:|
Ishola D A, Arogundade F A, Sanusi A A, Akinsola A. Association of Hydrocarbon Exposure with Glomerulonephritis in Nigerians: A Case Control Study. Saudi J Kidney Dis Transpl 2006;17:82-9
|How to cite this URL:|
Ishola D A, Arogundade F A, Sanusi A A, Akinsola A. Association of Hydrocarbon Exposure with Glomerulonephritis in Nigerians: A Case Control Study. Saudi J Kidney Dis Transpl [serial online] 2006 [cited 2020 Jan 23];17:82-9. Available from: http://www.sjkdt.org/text.asp?2006/17/1/82/32452
| Introduction|| |
Chronic renal failure (CRF) is a complex medical condition that presents major problems for clinical and public health management in developing countries including ours.,,,,,,, Since glomerulonephritis (GN) is Nigeria's leading cause of CRF, it is necessary to identify and investigate its putative etiological factors to determine to what extent they increase the risk., One such candidate etiologic factor is exposure to hydrocarbons (HC), which are commonly found in natural gas and petroleum products such as gasoline, kerosene, lubricating oils, waxes, solvents, varnishes, glues, cleaning fluids, dyes, polish removers, paints and aerosols.,
Experimental evidence for HC-induced glomerular disease includes the in vitro stimulation of mesangial cell proliferation by Benzo[a]pyrene (BaP), a widely-used polycyclic aromatic HC and observations of various forms of HC-induced glomerular damage in laboratory animals. ,, In humans, a variety of glomerular lesions have been reported in association with HC exposure, and renal dysfunction has been found to be more common in chronically HC-exposed workers than in the non exposed. ,,,, Clinical studies have compared exposure of HC in patients with GN and other renal disorders with control subjects. Though some investigators have found nonsignificantly increased exposure in patients relative to control groups,,, other reports have indicated that HC exposure levels were significantly higher in patients.,,,
The risk of glomerulonephritis due to the exposure to HC has not been previously studied in an African population. Interracial differences are well documented in many types of kidney disease and the finding of genetic influences in HC-associated renal damage specifically raises the possibility of racial variability in susceptibility to this effect.,
We have conducted this study to investigate the HC exposure as a possible risk factor for GN-induced CRF by comparing the estimated life-time exposure level in Nigerian patients with matched healthy controls.
| Patients and Methods|| |
We studied 95 subjects: 50 patients with CRF due to primary GN and 45 healthy controls matched for age and gender.
The patients were consecutively selected from the pool of CRF patients attending our nephrology clinic and/or admitted as inpatients. The defining criteria for CRF included serum creatinine consistently greater than 175 µmol/l and calculated creatinine clearance estimate of glomerular filtration rate (GFR) less than 60 ml/min. As renal biopsy was not routinely available, the diagnosis of primary GN as the cause of CRF was clinically determined using previously described clinical criteria in Nigerian patients by Akinsola et al.  These criteria included a documented past history of primary nephrotic syndrome or a verified long-standing history of frothy urine besides documented normotension and euglycemia prior to the onset of chronic renal disease. Subjects with sickle cell anemia, evidence of other antecedent systemic diseases, a history of recurrent urinary infections, or chronic analgesic use were excluded. Histological confirmation of GN as the cause of CRF was possible in 17 patients who consented to biopsy and had no contraindications to the procedure.
The matched control subjects were recruited at random from the Hospital's General Out-Patient Department. The screening process involved detailed interviews to exclude current and/or resolved symptoms that could suggest past or present renal disease, hypertension or diabetes. Appropriate physical examination and essential laboratory assessment including serum chemistry, urinalysis and urine microscopy were performed.
All the subjects in the study were residents in the same geographic region in Western Nigeria, in typically rural or semiurban, non-industrialized environments. They all signed informed consents. The study was approved by the institution's Ethics Committee.
Standard methods routinely used in our hospital laboratory were employed as appropriate to investigate the CRF patients and control subjects, including serum electrolytes, urea, creatinine, uric acid, and protein; urine microscopy; renal ultrasonography; fasting blood glucose; and hemoglobin electrophoresis. The patients underwent supervised 24-hour urine collection for creatinine clearance and urinary protein excretion measurements.
A hydrocarbon exposure assessment protocol was designed to determine the duration and intensity of HC exposure of each subject. The protocol was adapted from an existing scheme by Yaqoob et al  in the form of a questionnaire that gave appropriate relative weight to the intensity of exposure from different HC sources. Exposure to sources such as occupational house painting indoors, spray-painting without protection, carpet-cleaning agents, and production of paints and glues was regarded as heavy and allotted an intensity factor of 2. Activities such as spray-painting with protection, printing work, anesthetic work, and dry-cleaning were considered as moderate exposure and allotted an intensity factor of 1. Activities such as outdoor painting, motor repairing, and handling of petrol fuels were considered as mild expo-sure and allotted an intensity factor of 0.5. The total HC exposure of each individual was computed as the product of the total number of hours of exposure (to each exposure source) and the intensity factor(s) of the source(s). This resulted in a specific hydro-carbon exposure score (HES) for each subject, recorded in arbitrary absolute units.
The face and content validity of the questionnaire was established through scrutiny by the investigators and their colleagues during informal pre-study presentation and protocol review sessions. Pre-test questionnaire administration was conducted on randomly selected patients on our medical wards through the help of trained assistants (to avoid investigator bias); and the performance of the questions assessed in terms of ease of understanding by subjects and capacity to elicit the desired information. Through these procedures the questionnaire was progressively modified towards optimal sensitivity in estimating lifetime HC exposure in each subject, emphasizing the consideration of local sources of exposure. Inter-rater comparison was used to assess the reliability of the questionnaire, yielding reliability (concordance levels) between 0.7 and 0.8 - indicating good agreement according to the criteria of Fleiss . In the actual course of the study, test-retest "spot checks" (repeat questionnaire administration at least one week after initial interviews) showed a satisfactory agreement level of 0.8.
| Statistical methods|| |
For group data comparisons, the Student's t test (for normally distributed data), MannWhitney U test (for non-parametric data), and Chi-square (for proportions) were employed as appropriate. The HC exposure score data obtained had a non-parametric distribution. Logistic regression analysis was used to determine the significance of HC exposure in relation to other possible determinants of GN-induced CRF and to compute the odds ratio as an estimation of the strength of HC-GN association. Exposure scores were adapted into a percentile-based dichotomous variable to facilitate this. Differences were accepted as statistically significant a p <0.05. SPSS 10.0.5 (SPSS.Inc.) was used for statistical analysis.
| Results|| |
There were 50 subjects in the GN group (28 males (M), 22 females (F), M:F ratio 1.3:1); and 45 healthy control subjects (24 M, 21 F; M:F 1.1:1). [Table - 1] shows basic data from the two groups, including similar age and gender distribution. GN patients had mean 24-hour urinary protein excretion of 7.8 ± 5.5 (range 1.25 - 20.0) g/day and mean creatinine clearance of 24.3 ± 15.7 (range 2.0 - 55.0) ml/min.
The HC exposure scores (HES) were significantly higher in the GN patients (p<0.001, Mann-Whitney U test), [Table - 2]. The distribution pattern of individual HES within the two groups is shown in [Figure - 1]. We categorized all the study subjects into high exposure and low exposure sub-groups using the 67th percentile of all HES in the study as cut-off point. Thus, subjects with HES above the 67th percentile (i.e. upper third of HES) were considered to have high HC exposure while those with HES below this level were considered to have low HC exposure. [Figure - 2] shows the proportions of high or low exposure individuals in each group. There was significantly more highexposure indivi-duals in the GN group than controls (24 of 50 GN patients vs. 8 of 45 controls, p<0.002).
Logistic regression analysis was used to further evaluate the significance and strength of the relationship between HES and GN and to assess the influence of subjects' age and gender. A backward stepwise regression procedure based on likelihood ratios was applied to develop a model using GN group membership as the dependent variable. Covariates at the starting step were HES, gender and age. The HES was included in the model in a dichotomous variable form (high or low HC exposure as defined above) while age was included as a categorical variable (groups: <20 years, 20-29 years, 30-39 years, etc.). Stepwise elimination of gender and age from the model had nonsignificant effects (p = 0.8 and p = 0.6 respectively), resulting in a final model with high overall performance (-2 log likelihood 121.3; corresponding chi-square 10.0; p = 0.002). The coefficient of the HC exposure was 1.45 (p < 0.003), with a corresponding odds ratio of 4.3 (95% confidence interval 1.7 - 11) for GN-induced CRF from high HC exposure. Thus, high HC exposure was significantly associated with GN (more than four-fold increase in risk), but no age or gender effect was observed.
| Discussion|| |
This study explored the possibility of an association between GN-induced CRF and HC exposure in Nigerian patients. We found that the HC exposure was significantly higher in GN patients than in healthy controls irrespective of age and gender and increased the risk of GN-induced CRF more than four-fold. These findings support the hypothesis that the HC exposure may play a role in GN in Nigerian subjects and may have important implications for CRF prevention and management strategies. Decreased HC exposure may reduce the risk of developing GN. In the clinical evaluation of our GN and CRF patients, it may be beneficial to be more thorough in obtaining the history of occupational and recreational activities in order to caution against continuing HC exposure where applicable.
There were some limitations in our exposure assessment instrument. Direct quantification of the validity and specificity could not be obtained, as there is no "gold standard" method for measuring lifetime HC exposure. Also, the dependence of exposure assessment on recall from memory is not ideal. However, it is well recognized that in the risk assessment studies, exposure quantification is often a difficult and complex procedure, hence indirect measurements are frequently employed.  The questionnaire/recall method (as employed in this study) remains a vital tool in investigating such risk factors that lack specific laboratory-based measurements. We employed, as a surrogate marker for exposure, an arbitrary but standardized scoring system, which we adapted from an HC exposure assessment scheme that has been effectively used elsewhere. 
One of the difficulties of research into an association between HC exposure and renal disease is that different investigators have used widely varying measures of exposure. Hence our use of percentiles of exposure scores - and not just raw scores - in statistical analysis should be useful in easing comparison with studies of similar design irrespective of the differences in scoring systems employed.
There is a need for studies at both basic and clinical levels to further clarify the HCGN relationship. It is possible to design animal studies to maximize objectivity in exposure assessment through direct measurements of renal tissue HC content, which could then be related to the HC-induced renal pathology. The role of HC exposure not only in the pathogenesis but also in the progression of GN requires further assessment. As has been suggested by Ravnskov,  it is important to confirm whether GN patients who still possess good renal function would have a reduced risk of progression to CRF if the HC exposure was curtailed. Furthermore, it is possible that discontinuation of exposure could retard the rate of deterioration of renal function or result in reversal of progression if detected early; this is analogous to the regression of renal damage in diabetics when detected and treated at the stage of incipient diabetic nephropathy (isolated microalbuminuria). Other host and toxin factors, which appear to influence the effect of the HC on renal function, include genetic polymorphisms that exist in human HC metabolic biotransformation pathways. They could be important determinants of individual or racial susceptibility to HC-induction renal damage. Moreover, the previously observed variation in renal toxicity among various types of HC compounds indicates that the occurrence or degree of renal damage may depend on the specific HC substance involved in the exposure.,
The mechanism of HC-induced renal damage is not yet clear but is thought to be a consequence of HC tubular toxicity. It has been postulated that constant low-grade tubular damage due to chronic HC exposure results in the release of sequestered or altered tubular antigens, which then result in GN by provoking autoimmunity. Pai et al  suggested that since HC catabolism generally depends on mixed function oxidase systems that generate alcohols, aldehydes and organic acids, reactive oxidative species or intermediates within these processes may cause renal tubular damage and subsequent glomerular injury. These hypotheses need to be further investigated.
In conclusion, our study demonstrates an association between GN-induced CRF and HC exposure in Nigerian patients. We suggest further investigation of the risks and preventive strategies to limit the occurrence of GN, a major cause of end-stage renal disease in Nigerians.
| Acknowledgement|| |
This work was partly supported by a Research Grant to DAI from the Obafemi Awolowo University Teaching Hospital, Nigeria.
| References|| |
|1.||Plange-Rhule J. Chronic renal failure: practical difficulties in management. Africa Health 1996;18 (3):15-6. |
|2.||Akinkugbe OO. Tropical nephropathy-an overview. Afr J of Med Sci 1992;21:3-7. |
|3.||Akinkugbe OO. Nephrology in the tropical setting. Nephron 1978;22:249-52. [PUBMED] |
|4.||Adelekun TA, Akinsola A. Hypertension induced chronic renal failure: clinical features, management and prognosis. West Afr J Med 1998;17:104-8. [PUBMED] |
|5.||Fogazzi GB, Attolou V, Kadri S, Fenili D, Priuli F. A nephrological program in Benin and Togo. Kidney Int Suppl 2003;83:S56-60. |
|6.||Naicker S. End-stage renal disease in sub-Saharan and South Africa. Kidney Int Suppl 2003; 83 :S119-22. [PUBMED] |
|7.||Akinsola A, Adelekun TA, Arogundade FA. CAPD practice in OAUTH, Ile-Ife, Nigeria: A preliminary experience. Dialysis and Transplantation 2000; 29:774-82. |
|8.||Bamgboye EL. Hemodialysis: management problems in developing countries, with Nigeria as a surrogate. Kidney Int Suppl 2003; 83: S93-5. [PUBMED] |
|9.||Ojo OS, Akinsola AA, Nwosu SO, Odesanmi WO. The patho-logical basis of chronic renal failure in Nigerians. An autopsy study. Trop Geog Med 1992;44:42-6. |
|10.||Akinsola W, Odesanmi WO, Ogunniyi JO, Ladipo GO. Diseases causing chronic renal failure in Nigerians-a prospective study of 100 cases. Afr J Med Sci 1989;18:131-7. |
|11.||Housecroft CE, Constable EC (1997). Alkanes, Alkenes and Alkynes. In Chemistry: An Integrated Approach. (eds. CE Housecroft, EC Constable) Addison Wesley Longman Limited, Harlow;331-91 |
|12.||Vale JA, Meredith TJ. Poisoning from hydrocarbons, solvents and other inhalational agents. In Oxford Textbook of Medicine. 1st edn. (eds. DJ Weatherall, JGG Ledingham, DA Warrell) Oxford University Press; Oxford:6.27-6.33. |
|13.||Bowes RC 3rd, Weber TJ, Ramos KS. Induction of highly proliferative phenotypes in cultured glomerular mesangial cells by benzo [a]pyrene alone or in combination with methoxamine. Arch Biochem Biophys 1995;323:243-50. [PUBMED] [FULLTEXT]|
|14.||Zimmermann SW, Norbach DH. Nephrotoxic effects of long-term carbon tetrachloride administration in rats. Arch Pathol Lab Med 1990; 104:94-9. |
|15.||Mattie DR, Alden CL, Newell TK, Gaworski CL, Flemming CD A 90-day continuous vapor inhalation toxicity study of JP-8 jet fuel followed by 20 or 21 months of recovery in Fischer 344 rats and C57BL/6 mice. Toxicol Pathol 1991;19:77-87. |
|16.||Ehrenreich T, Yunis SL, Churg J. Membranous nephropathy following exposure to volatile hydrodcarbons. Envir Res 1997;14:35-45. |
|17.||Kleinknecht D, Morel-Maroger L, Callard P, Adhemar JP, Mahieu P. Antiglomerular basement membrane nephritis after solvent exposure. Arch Intern Med 1980;140:230-2. [PUBMED] |
|18.||Klavis G, Drommer W. Goodpasture's syndrome and the effects of benzene. Arch Toxikol 1970;26:40-55. [PUBMED] |
|19.||Viau C, Bernard A, Lauwerys R, et al. A cross sectional survey of kidney function in oil refinery employees. II, p Am J Ind Med 1987;11:177-87. |
|20.||Yaqoob M, Bell GM, Stevenson A, Mason H, Percy DF. Renal impairment with chronic hydrocarbon exposure. Q J of Med 1993;86:165-74. |
|21.||Harrington JM, Whitby H, Gray CN, Reid FJ, Aw TC, Waterhouse JA. Renal disease and occupational exposure to organic solvents: a case referent approach. Br J Ind Med 1989;46:643-50. [PUBMED] |
|22.||van der Laan G. Chronic glomerulonephritis and organic solvents. A casecontrol study. Int Arch Occup Environ Health 1980;47:1-8 |
|23.||Stengel B, Cenee S, Limaset J-C, Protois JC, et al. Organic solvent exposure may increase the risk of glomerlar nephropathies with chronic renal failure. Int J Epidemiol 1995;24:427-34 |
|24.||Yaqoob M, Bell GM, Percy DF, Finn R. Primary glomerulonephritis and hydrocarbon exposure: a case-control study and literature review. Q J Med 1992;83:(301)409-18. |
|25.||Ravnskov U. Exposure to organic solvents - a missing link in post- streptococcal glomerulonephritis. Acta Med Scand 1978;203:351-6. [PUBMED] |
|26.||Zimmerman SW, Groehler K, Beirne GJ. Hydrocarbon exposure and chronic glomerulonephritis. Lancet1975;2:199-201. [PUBMED] |
|27.||Harrison DJ, Thomson D, MacDonald MK. Membranous glomerulonephritis. J Clin Pathol 1986;39:167-7. [PUBMED] [FULLTEXT]|
|28.||Bowden DW. Genetics of kidney disease. Kidney Int Suppl 2003;63:S8-12. |
|29.||Pai P, Hindell P, Stevenson A, Mason H, Bell GM. Genetic variants of microsomal metabolism and susceptibility to hydrocarbon-associated glomerulonephritis. Q J of Med 1997;90:693-8. |
|30.||Fleiss JL. The measurement of interrater agreement. In Statistical methods for rates and proportions (ed. JL Fleiss), 2nd ed., John Wiley, New York 1981;: 211-36. |
|31.||Wakefield J. Human exposure: the key to better risk assessment. Environ Health Perspect 2000;108:A559-65. [PUBMED] [FULLTEXT]|
|32.||Ravnskov U. Influence of hydrocarbon exposure on the course of glomerulonephritis. Nephron 1986;42:156-60. [PUBMED] |
D A Ishola
FWACP Renal unit, Department of Medicine, Obafemi Awolowo University Teaching Hospital, Ile-lfe
[Figure - 1], [Figure - 2]
[Table - 1], [Table - 2]