This review is limited to studies involving human subjects. Animal studies have not been included. Several different approaches have been adopted in the investigation of the effects of PUFA ingestion on the risk of nephrolithiasis in humans. Approaches include stone incidence rates, baseline habitual diets, baseline urinary and plasma PUFA profiles, and the effects of PUFA dietary interventions on urinary and plasma risk factors in stone formers and healthy subjects.
2.1. Population Studies of Stone Incidence
Researchers have investigated population groups in which stone incidence is rare and whose habitual dietary practices with respect to PUFA ingestion are documented. Two such population groups have been identified during the past 70 years—the Inuit peoples of the Arctic regions of Greenland, Canada, and Alaska (formerly referred to as Eskimos) and the black inhabitants of South Africa (formerly referred to as Bantu). As the following summary will indicate, it is somewhat ironic that although some of these population-based studies are founded on a self-perpetuating urban legend and other studies have reported inconclusive or counterintuitive findings, they have nevertheless generated high levels of interest which themselves have led to excellent research efforts finally culminating in the present review.
The notion that PUFAs occurring in fish oils is protective against renal stone disease arises from the widely cited claim that urolithiasis is extremely rare in Alaskan (Canadian) and Greenland Eskimos, whose diet is predominantly rich in this food source. However, scrutiny of the mainstream literature reveals that the observation of a protective effect against kidney stones per se does not seem to have been reported. The earliest mention of stone rarity in Eskimos that we could find was reported in the paper by Modlin [16
] in which he states, “Eskimos rarely form renal stones”, but he does not provide any literature references to substantiate this point. Interestingly, in a paper on the relative importance of essential fatty acids in the Eskimo diet published 14 years later, Sinclair provides a list of several diseases that are rare in Eskimos [17
]. Although he includes gallstones, he does not mention renal stones. Ironically, several studies by other authors routinely cite these papers (particularly the one published by Modlin) as the authoritative references for the epidemiological anomaly in Eskimos [1
]. Other studies also refer to the rarity of renal stones in Eskimos but mistakenly provide references which describe the rarity of heart atherosclerotic and degenerative diseases without mentioning renal stones [4
]. What is undisputed, however, is that the habitual diet of Eskimos is rich in n-3 PUFAs [17
] and that the occurrence of several degenerative diseases is extremely rare in this population group [1
]. This epidemiological observation, coupled with the notion of an apparent rarity of stones, has motivated many excellent research studies to investigate the role of PUFAs in stone disease.
The rarity of renal stones in the South African black population (B) (<1%) accompanied by a corresponding incidence rate of 10% in the white population (W) is well documented [21
]. In a recent study of the FA content in the habitual diets of 10 white and 10 black South African subjects, the authors hypothesized that potential differences in the respective dietary PUFA intakes in the groups might demonstrate a protective effect in B [15
]. However, their findings did not support their hypothesis as differences in the intakes of three PUFAs, important in the assessment of stone risk, and their concomitant concentrations in plasma and red blood cell total phospholipids were contrary to what might have been expected. Somewhat counterintuitively, the findings indicated a higher risk in the B group [15
]. Details of the mechanisms by which PUFAs affect the risk factors for stone formation are given in the following paragraphs.
2.3. Baseline Phospholipid PUFA Profiles in Stone-Formers vs. Healthy Subjects, and in High-Risk vs. Low-Risk Groups
To test the notion that PUFAs may play a role in the prevention or pathogenesis of renal stone formation, researchers have speculated that fundamental differences might exist in PUFA profiles in stone-formers and healthy individuals. Attention has focused on n-3 FAs like alpha-linolenic acid (ALA), and EPA, and on n-6 FAs like linoleic acid (LA), gamma-linolenic acid (GLA), and arachidonic acid (AA). Baggio and co-workers tested the hypothesis that ion flux cell abnormalities (previously associated with calcium oxalate stone formation) were secondary effects to an anomaly in renal cell membrane composition [11
]. They found lower content of LA and a higher concentration of AA in plasma and erythrocyte membrane phospholipids and an increased AA/LA ratio in idiopathic calcium stone-formers compared to healthy controls [11
]. A higher concentration of AA in stone formers was confirmed in a later study by Baggio et al., who also reported higher PGE2
in this group [9
]. In the same year, Messa et al. performed a study to confirm these findings and to test whether any relationship exists between the FA composition of red blood cell membranes and the main metabolic factors involved in stone formation [12
]. Interestingly, in direct contradiction to the studies of Baggio et al., they reported lower AA, LA, and DHA in the red blood cell membranes of stone-formers [12
]. However, they did find that hyperoxaluric stone-formers had a relatively higher AA than stone-formers with normal oxalate excretion [12
]. Finally, Rodgers et al. argued that the significantly lower occurrence of stones in South Africa’s black population relative to that in the white population presented them with an ideal opportunity to test whether PUFAs might play a role in this anomaly. They examined PUFA plasma and urinary profiles in their low- and high-risk groups B and W, respectively [15
], and reported findings that were also counterintuitive relative to the Baggio model [11
]. The concentration of AA was significantly higher in the low-risk B group, and there was no difference between the groups in AA/LA ratios [15
]. Thus, consideration of the findings reported in this paragraph leads to the conclusion that inter-group comparisons in PUFA profiles have not been consistent, and, as such, have not provided irrefutable evidence of their potential role in nephrolithiasis.
2.4. PUFA Dietary Interventions
Numerous intervention studies have been conducted. Researchers have focused their attention on examining the effects (if any) of PUFA ingestion on the well-established physicochemical risk factors of urinary calcium, oxalate, citrate, magnesium, and phosphate [26
]. A summary of these is given in Table 1
. We note that a total of 16 studies is listed. PUFAs which were tested are EPA (2 studies) [1
], EPA+DHA (8 studies) [1
], LA+GLA (5 studies) [2
], and EPA+DHA+GLA+LA (one study) [2
]. These have been delivered as fish oil or as isolated supplements (EPA, GLA, evening primrose oil).
Urinary Ca excretion decreased in 10 of these (62.5%). There was no change in this urinary parameter in the remaining 6 studies. Inspection of the composition of the test substances shows that EPA and DHA are common in 8 of the 10 studies in which urinary Ca excretion decreased, while LA and GLA (as evening primrose oil) account for the other two.
In examining the efficacy of PUFAs for reducing urinary Ca, we need to account for the studies in which this urinary parameter did not change. In their paper on the effect of EPA, Yasui and co-workers found a reduction in stone recurrence [5
]. Surprisingly urinary Ca excretion was unchanged [5
]. This was in contrast an earlier study of theirs in which the same EPA supplement was used but urinary Ca decreased [4
]. They suggested that this inconsistency might be due to differences in their respective patient cohorts as their earlier study involved hypercalciuric patients while their subsequent study did not [5
]. Interestingly, Table 1
shows that four of the other studies in which no decrease in urinary Ca occurred used healthy subjects as their test group as opposed to stone-formers [6
] lending support to the notion that PUFA-induced reduction of Ca excretion is possibly restricted to stone-forming patients. Indeed, Buck and co-workers commented in an earlier paper that the normalizing effect of fish oil with respect to urinary calcium and oxalate levels was more pronounced in patients in whom these parameters were markedly raised [1
]. The table also shows that evening primrose oil (i.e., LA+GLA) achieved a reduction in Ca excretion in healthy subjects [10
] but not in stone-formers [2
]. Clearly, the number of studies involving evening primrose oil is too limited to warrant us pronouncing on its efficacy.
shows that urinary oxalate was measured in thirteen studies after the administration of the PUFA test substance. It decreased in four of these [1
] but remained unchanged in eight studies [2
]. (Reference 2 describes three studies; reference 10 describes 2 studies; reference 25 also describes 2 studies, but an unchanged urinary oxalate was recorded in only one of them). One study found that it increased [25
]. Of importance is that the supplement that was administered in the four studies in which urinary Ox decreased was fish oil. Other supplements (LA/GLA and GLA alone) [2
] had no effect on Ox excretion. Fish oil failed to lower Ox excretion in three studies [2
]. Rothwell et al. commented that this difference in the effect of fish oil on oxalate handling is difficult to explain but that relatively higher baseline Ox levels may respond more readily to PUFA supplementation [8
]. This is in accordance with the previously mentioned comment by Buck and co-workers that the normalizing effect of fish oil is more pronounced in patients with relatively higher Ox excretion levels [1
]. This was subsequently confirmed by Lange et al. [14
Urinary citrate was measured in 10 studies involving different PUFAs. Five of these were performed after fish oil ingestion [2
], three were performed after EPO ingestion (2,10), and two were performed after GLA ingestion [25
] (references 10 and 25 each describe studies invoving two race groups, B and W). Although an increase in urinary citrate was observed in 40% of these [7
] (both race groups in reference 10; W race group in reference 25), no change in this parameter occurred in any of the other studies [2
] (B race group in reference 25) except for one study in which a decrease was observed [8
]. Clearly, the effect of PUFAs on the urinary excretion of citrate is yet to be fully established, and as such, it requires further investigation in future studies. Similarly, changes in urinary Mg and Phos excretion have occurred too seldom for us to draw any firm conclusions.
Of critical importance is that there has been only one study that has used stone recurrence as its outcome measure for assessing the efficacy of PUFAs in reducing stone formation per se [5
]. In that study, 29 CaOx stone-forming patients were followed over a period of 8 years, during which a highly purified supplement of EPA was administered (1.8 g per day for three years). Stone recurrence was monitored before, during, and after supplementation. The incidence rate of nephrolithiasis was significantly lower during the administration of the supplement compared to before and after its administration. Intriguingly, no change was observed in any of the urinary parameters that were measured, including Ca, Mg, and Phos, thereby ruling out the possibility of EPA having reduced stone incidence via alterations in these urinary risk factors. Remarkably, urinary Ox was not measured in this study [5
]. We can only speculate that a reduction in the urinary excretion of this component might have been a contributory factor in reducing stone incidence. Of course, the possibility exists that some factor besides urine biochemistry might have caused this effect. Future studies involving the effect of PUFA ingestion on stone recurrence will most likely resolve this puzzle.
Consideration of the findings reported in Table 1
and described in the preceding paragraphs indicates that the PUFAs EPA and DHA (either in combination in fish oil or as individual, purified supplements) reduced urinary Ca in 66.7% and urinary oxalate in 57.1% of the studies in which they were measured. In the remainder of these studies (33.3% and 42.9%, respectively), there was no change in either parameter. Findings also indicate that a favorable outcome involving an increase in urinary citrate was observed in four studies involving different PUFAs [7
] (both race groups in reference 10; W race group in reference 25). Since these findings are potentially promising for possible application in stone management, explanatory mechanisms for these effects are warranted.