Low-carbohydrate–high-protein diets

Is there a place for them in clinical cardiology?

What is an LC-HP diet?

Usually, LC-HP diets are those that contain significant quantities of animal protein and relatively low amounts of carbohydrates, rendering them ketogenic. Individuals who consume such diets are in a perpetual state of ketosis, which leads to a disproportionate use of fat stores for energy. This is the normal response to caloric deprivation. Table 1compares two LC-HP diets (5,6)currently in vogue, with conventional recommendations with which physicians are familiar (3,7). It is clear that in the LC-HP diets, the percentage of daily calories provided by protein, total fat, and saturated fat is 2 to 2.5 times higher than in the AHA guidelines. The corresponding value for carbohydrates is typically 30% to 90% lower. Where do these diets, with a surfeit of saturated fat, cholesterol, and practically no carbohydrate and fiber, fit into the management of patients? Are they a weight-loss regimen, or have they other therapeutic benefits for individuals with cardiovascular disease?

LC-HP diets and weight loss

A basic tenet of a dietary approach to weight loss is caloric restriction. In contrast, with LC-HP diets, caloric restriction is not imposed but appears to be an inevitable outcome (8,9). Skov et al. (9)showed that the resulting weight loss on high-protein diets continued for as long as the subject remained on the study protocol (up to six months). These authors concluded that high-protein diets were an effective means of reducing caloric intake and speculated that it was probably due to appetite suppression secondary to ketosis. Regardless of the potential mechanisms involved, there is little doubt that an LC-HP diet results in weight loss. Although similar changes are observed with more conventional low-fat dietary regimens, they have been criticized on the basis that the latter simply “do not work” and do not extend life expectancy (10). Surprisingly, the LC-HP diets have escaped this criticism, presumably because of the widely held but mistaken belief that they are effective invariably. Nevertheless, there is a considerable body of evidence to support the claim that a low-fat diet available ad libitum results in a significant weight loss even when it is the sole intervention(11,12). Additionally, when tested in a prospective fashion, the attrition (i.e., failure) rate in healthy individuals on LC-HP diets varies from 20% to 43% in the short term (<6 months) (13–16). These attrition rates are similar to those observed with more conventional weight-loss regimens (14–16). Indeed, when an LC-HP diet is compared directly with other dietary regimens for weight loss, it appears that the weight loss is not a unique consequence of the LC-HP nature of the diet but rather a function of the reduced caloric intake (14–16).

The recent study reported by Westman et al. (1), which compares an Atkins-type diet with a low-fat diet provides some interesting insights into the relationship between calorie restriction and weight loss. At the time of this writing, the caloric contents of the two diets in the study have not been published. The mean age of subjects was 46 years; the mean baseline body mass index (BMI) was 34.5 kg/m². The weight loss over six months was 13.8% for the Atkins-type diet (n = 36) and 8.8% for the low-fat diet (n = 27). However, in a previous publication Westman et al. (13)reported in greater detail the effects of a similar LC-HP diet in a group of 51 free-living individuals motivated to lose weight. Of the individuals, 41 completed 24 weeks on the diet. While participating in the study, they were instructed to eat unrestricted quantities of beef, pork, chicken, fish, shellfish, and eggs. Cheese intake was permitted to 4 oz/day. They were also asked to eat two cups of salad vegetables and one cup of a low-carbohydrate vegetable. At the start of the study, the subjects consumed 25 g of carbohydrates per day until the target weight was attained and thereafter increased carbohydrate consumption to 40 g/day. Their initial BMI ranged from 26 to 33 kg/m². After 24 weeks on the diet, there was an average weight loss of 19.8 lbs in the 41 subjects, representing an average reduction in BMI of 3.2 kg/m². Reductions in serum cholesterol, LDL, and triglycerides were observed together with a concurrent increase in high-density lipoprotein. In both of these studies, all of the patients were overweight and were motivated to lose weight.

A closer examination of the data presented in the foregoing study provides a relatively simple explanation for the observed weight loss. The most important consideration in this study (13)is the daily consumption of calories. Despite the unrestricted nature of the diet, the total energy consumption as assessed from food diaries was only 1,447 calories/day. It is of interest to compare this value with the estimates of energy requirements based on the new guidelines proposed by the Institute of Medicine/National Academy of Science (17). From the data presented in Table 1of Westman et al. (13), one could estimate the energyexpenditure to be approximately 2,775 calories for the men and 2,313 calories for the women in the study (17). It is widely recognized that self-reported food diaries underestimate the calorie intake by 10% to 20% if they are used without an appropriate correction term (18,19). Even after applying such a correction, it would appear these subjects endured a significant calorie deficit during the 24 weeks of the study. It is likely that this deficit was at least 400 calories/day. Because it is generally recognized that a cumulative deficit of 3,500 calories would result in a weight loss of 1 lb, one could anticipate a weight loss of approximately 19.2 lbs (i.e., [400 × 7 × 24]/3,500) over 24 weeks. In fact, the subjects lost an average of 19.8 lbs. Thus, simple calorie restriction alone could account for the weight loss observed on this LC-HP diet. There is also evidence to support the suggestion that a significant portion of the initial weight loss is due to loss of body water (20).

LC-HP diet and fasting lipid profile

In addition to weight loss, another claim made on behalf of the LC-HP diet is its potentially favorable impact upon serum lipid profiles. In the study reported by Westman et al. (13), there was a reduction in total, LDL, and very-LDL cholesterol. It also showed that, in addition to reducing total LDL cholesterol levels, the LC-HP diet was associated with a reduction in the percentage of small LDL particles. Sharman et al. (21)had reported a similar trend in LDL particle size even in the absence of weight loss with a ketogenic diet. Recent studies reported do not confirm these favorable trends (15,16). In these studies, an LC-HP did not result in a significant reduction in either total or LDL cholesterol concentrations after six months. The high-density lipoprotein cholesterol concentration increased in one (15)but not in the other (16). However, there was a significant reduction in triglycerides in individuals who consumed a low-carbohydrate diet in both studies. Nevertheless, it should be noted that many of these favorable changes, including a reduction in serum LDL, could be achieved with conventional weight loss diets either alone (22)or in combination with exercise (23,24). Further, low-fat, high-carbohydrate diets do not consistently decrease LDL particle size (25). An additional issue that merits consideration is the effect of repeated high-fat meals on lipids in the postprandial state.

What practical advice should be given to individuals who seek information about the LC-HP diets?

Given the media focus of LC-HP diets, it is inevitable that physicians will encounter patients who insist on embarking on this type of diet. It is important that these patients understand that the long-term (beyond six months) consequences of an LC-HP, hypocaloric diet are unknown and that they have a clear appreciation of the adverse effects that could be expected while on the diet (Table 2).

In conclusion, use of LC-HP diets run counter to all the current evidence-based dietary recommendations for healthy populations (3,17). These diets do not meet the nutritional requirements of healthy people based on the current dietary reference intakes for many vitamins and minerals and recommendations for dietary fiber. When used for weight loss, these diets are associated with several potential adverse effects and nutrient deficits, and the long-term consequences of their continued use are unknown. On the basis of evidence currently available, LC-HP diets cannot be recommended as a part of a long-term care plan for weight management in patients who smoke or have common diseases that affect the cardiovascular system, such as hypertension, hyperlipidemia, diabetes mellitus, and coronary atherosclerotic vascular disease, where endothelial dysfunction is a feature.