Literature search results and overview of included studies

The three literature searches resulted in the identification of 1697 titles, of which eighteen publications met all of the inclusion criteria and none of the exclusion criteria (Fig. 1).

Fig. 1. Flowchart of the literature search process.

The eighteen publications provided a total of twenty-seven strata (Table 1). Of the twenty-seven strata, seventeen were from parallel trials, and the remaining ten strata were from crossover trials. The number of study completers ranged from thirteen to 137 amongst the eighteen publications. Both male and female subjects were studied in the majority of the strata, with the exception of three strata wherein only females were studied (Abazarfard et al. ⁽ Reference Abazarfard, Salehi and Keshavarzi ^{9 )}; Kurlandsky & Stote strata 1 and 2⁽ Reference Kurlandsky and Stote ^{14 )}) and two strata wherein only males were studied (Jia et al. strata 1 and 2⁽ Reference Jia, Li and Zhang ^{15 )}). The subjects were described by the authors as generally healthy in seven strata (Abazarfard et al. ⁽ Reference Abazarfard, Salehi and Keshavarzi ^{9 )}; Berryman et al. ⁽ Reference Berryman, West and Fleming ^{11 )}; Foster et al. ⁽ Reference Foster, Shantz and Vander Veur ^{16 )}; Kurlandsky & Stote strata 1 and 2⁽ Reference Kurlandsky and Stote ^{14 )}; Spiller et al. strata 1 and 2⁽ Reference Spiller, Jenkins and Bosello ^{17 )}), generally healthy but habitual smokers in two strata (Jia et al. strata 1 and 2⁽ Reference Jia, Li and Zhang ^{15 )}), generally healthy or hyperlipidaemic in two strata (Sabaté et al. strata 1 and 2⁽ Reference Sabaté, Haddad and Tanzman ^{18 )}), hyperlipidaemic in four strata (Damasceno et al. ⁽ Reference Damasceno, Perez-Heras and Serra ^{19 )}; Jenkins et al. strata 1 and 2⁽ Reference Jenkins, Kendall and Marchie ^{20 )}; Tamizifar et al. ⁽ Reference Tamizifar, Rismankarzadeh and Vosoughi ^{21 )}), or pre-diabetic or at risk of type 2 diabetes mellitus (T2DM) in five strata (Tan & Mattes strata 1 to 4⁽ Reference Tan and Mattes ^{22 )}; Wien et al. ⁽ Reference Wien, Bleich and Raghuwanshi ^{23 )}). In the remaining seven strata, medicated subjects were studied, including subjects with T2DM on oral hypoglycaemic therapy (Sweazea et al. ⁽ Reference Sweazea, Johnston and Ricklefs ^{10 )}; Cohen & Johnston⁽ Reference Cohen and Johnston ^{24 )}; Li et al. ⁽ Reference Li, Liu and Liu ^{25 )}; Lovejoy et al. strata 1 and 2⁽ Reference Lovejoy, Most and Lefevre ^{26 )}), subjects on stable statin therapy (Ruisinger et al. ⁽ Reference Ruisinger, Gibson and Backes ^{27 )}), and subjects on antihypertensive and/or antidiabetic medications (Wien et al. ⁽ Reference Wien, Sabaté and Iklé ^{28 )}).

Table 1. Key study characteristics of included studies (n 18 publications and 27 strata)

R, randomised; C, controlled; P, parallel; F, females; X, crossover; WO, washout; M, males; T2DM, type 2 diabetes mellitus; SB, single-blinded; NR, not reported; TDEI, total daily energy intake; DB, double-blinded; NCEP ATP III, National Cholesterol Education Program Adult Treatment Panel's Third Report on Therapeutic Lifestyle Changes; CHO, carbohydrates; ACE, angiotensin-converting enzyme; HRT, hormone replacement therapy; ADA, American Diabetes Association.

* For P studies, the weighted mean age/BMI and pooled sd were calculated by using the mean age/BMI and sd values that were reported separately for the individual groups.

† Subjects had not been prescribed insulin; rather, subjects were on stable oral hypoglycaemic therapy.

‡ In another arm of this trial, subjects were administered 40 to 65 g/d of walnuts; however, according to the study exclusion criteria, the control is not acceptable if it consists of another tree nut or any tree nut fraction. Thus, the results of the walnuts intervention arm were excluded from the analysis.

§ Damasceno et al. ⁽ Reference Damasceno, Perez-Heras and Serra ^{19 )} acknowledged that, given the nature of the foods provided, which could not be masked, the study was unblinded; however, it was noted that investigators involved in the preparation of databases and laboratory determinations were masked with respect to treatment sequence.

║ The almonds partially replaced other MUFA-rich foods, such as olives and avocados.

¶ The Mediterranean-type, cholesterol-lowering diet was composed of natural foodstuffs. Vegetable products and fish were emphasised, while red and processed meats, whole-fat dairy products and eggs were limited. Recipes using nuts were provided to the subjects who consumed the nuts with meals in desserts or salads or as snacks.

** During the first 5 weeks, subjects received whole, raw almonds only. From week 6, roasted almonds were introduced and, over time, a variety of isoenergetic, flavoured almonds were used.

†† The majority of the subjects were not medicated; however, 3 M and 5 F were taking the following medications: statins (n 2), β-blocking agents (n 3), ACE inhibitors (n 3), angiotensin II AT1 receptor blockers (n 1), thiazide diuretics (n 2), levothyroxine (n 2) and HRT (n 2). Medication dosages were held constant throughout the study.

‡‡ Subjects did not use lipid-lowering medications or dietary supplements. Subjects were allowed to be on stable regimens of oral contraceptives and HRT.

§§ Subjects did not receive insulin therapy; rather, all subjects were on stable oral hypoglycaemic therapy.

║║ Almonds were incorporated into the control diet to replace 20 % of TDEI in the control diet; depending on the menus, almonds were either incorporated into entrées and desserts or consumed as a snack.

¶¶ In this study, subjects were provided with all foods needed for the duration of the study; since the study was crossover in design but described as DB, it is assumed that almond powder was used in the formulation of the entrées and snacks. However, it remains unclear how true double-blinding was achieved, given that the subjects could have tasted the almonds in their foods (it is possible that the ‘control’ foods were flavoured with an almond extract; however, this was not described in the publication).

*** Subjects on lipid-lowering or insulin therapy were excluded. Eleven F received HRT, and sixteen of the subjects were taking oral hypoglycaemic agents.

††† Regarding background diets, on weekdays, the subjects were required to consume breakfast and dinner under supervision at the Pennington Biomedical Research Center's dining facility; weekday lunches and snacks and all weekend meals were packaged for take-out.

‡‡‡ Subjects were taking chronic statin therapy, defined as a consistent statin dose for at least 8 weeks before study entry with continuation of the same dose during the 4-week study period. Subjects who took lipid-lowering agents other than statins were excluded. Post-menopausal F who were not taking HRT or were on a consistent HRT dose were included. F of child-bearing potential using an effective form of contraception were allowed to participate in the study.

§§§ Subjects received NCEP ATP III diet counselling via telephone and instructions on how to compensate for the added energy from the almonds.

║║║ The authors argued that a WO was not included because serum lipids and lipoproteins are known to stabilise within 3 weeks.

¶¶¶ Although the mean BMI was NR, it should be noted that a BMI >30 kg/m² was an exclusion criterion.

**** On Sunday to Friday of each week, subjects ate breakfast and dinner at the Loma Linda University metabolic kitchen. Lunch meals and all Saturday meals were packaged for consumption away from the metabolic kitchen.

†††† The mean age was provided for the forty-five study completers. Because the mean age was not provided for each group, the weighted mean age for each comparison could not be calculated.

‡‡‡‡ All subjects were provided with whole-grain bread, brown rice, pasta, non-fat yogurt, rice cakes, dry beans, lentils, and couscous and were instructed to eat these foods a set number of times during each week. Subjects rounded out their daily food intake with fruits, vegetables, other whole grains, legumes, low-fat or non-fat milk (whole milk dairy products were not allowed), egg whites and lean fish. Lean beef was allowed twice weekly, and poultry and fatty fish were permitted up to four times weekly. Up to four whole eggs were permitted/week, but only if the subject had been consuming eggs prior to the study. Foods not allowed included: commercial or homemade products containing fats other than the study fat, and products made with refined flour (e.g. snack foods, chips, crackers, cakes, pastries, pies, candy or ice cream). Usual coffee, tea, alcohol and soft drink consumption was permitted.

§§§§ Subjects on insulin therapy were excluded. Subjects taking prescription medications, including oral hypoglycaemic agents, statins or hypertensive medications were instructed to maintain consistent use throughout the study.

║║║║ The study by Tamizifar et al. ⁽ Reference Tamizifar, Rismankarzadeh and Vosoughi ^{21 )} was described as SB. Given that, during the almond phase of the crossover study, the subjects were given almond powder and, during the control phase, the subjects were not administered anything, it is assumed that the investigators were blinded, and not the subjects.

¶¶¶¶ Subjects were considered at risk of T2DM if they were overweight or obese (BMI >27 kg/m²) or were normal weight (BMI 18·5–24·9 kg/m²) but had a strong family history for T2DM.

***** The proportion of subjects with a diagnosis of hypertension at baseline was 62 % in the control group and 50 % in the almond group; the proportion of subjects with T2DM was NR. However, it was noted that during randomisation, subjects were stratified according to the presence or absence of T2DM.

††††† Prediabetes was diagnosed according to the 2005 ADA diagnostic guidelines: fasting blood glucose between 100 and 125 mg/dl (5·56 and 6·94 mmol/l) or casual blood glucose ≥140–199 mg/dl (≥7·78–11·06 mmol/l).

‡‡‡‡‡ Subjects taking corticosteroids or immunosuppressant medications were excluded. Two subjects in each group were taking lipid-lowering medications.

Almond interventions

Across all strata, the average daily intake of almonds ranged from 20 to 113 g/d, and the duration of the almond consumption period ranged from 4 weeks to 18 months. Almonds were required to be consumed every day in all studies except two, in which 28 g (1 oz) of almonds were required to be consumed 5 d per week⁽ Reference Cohen and Johnston ^{24 )} or 43 g (1·5 oz) of almonds were required to be consumed five to seven times weekly⁽ Reference Sweazea, Johnston and Ricklefs ^{10 )}.

Whole, raw (unblanched, unsalted) almonds were consumed in nine strata (Abazarfard et al. ⁽ Reference Abazarfard, Salehi and Keshavarzi ^{9 )}; Sweazea et al. ⁽ Reference Sweazea, Johnston and Ricklefs ^{10 )}; Damasceno et al. ⁽ Reference Damasceno, Perez-Heras and Serra ^{19 )}; Jenkins et al. strata 1 and 2⁽ Reference Jenkins, Kendall and Marchie ^{20 )}; Ruisinger et al. ⁽ Reference Ruisinger, Gibson and Backes ^{27 )}; Spiller et al. strata 1 and 2⁽ Reference Spiller, Jenkins and Bosello ^{17 )}; Wien et al. ⁽ Reference Wien, Sabaté and Iklé ^{28 )}). In five strata (Cohen & Johnston⁽ Reference Cohen and Johnston ^{24 )}; Tan & Mattes strata 1 to 4⁽ Reference Tan and Mattes ^{22 )}), the almonds that were consumed by the subjects were not specifically described by the study authors as raw, unblanched almonds; however, based on the reported energy value of the almonds, it was determined that the almonds were most probably raw, unblanched almonds. A variety of almonds, namely whole (raw), roasted, and flavoured almonds were consumed in one stratum (Foster et al. ⁽ Reference Foster, Shantz and Vander Veur ^{16 )}); dry, roasted almonds were consumed in one stratum (Wien et al. ⁽ Reference Wien, Bleich and Raghuwanshi ^{23 )}); and almond powder was consumed in one stratum (Tamizifar et al. ⁽ Reference Tamizifar, Rismankarzadeh and Vosoughi ^{21 )}). The types of almonds used were not specified in two strata (Kurlandsky & Stote strata 1 and 2⁽ Reference Kurlandsky and Stote ^{14 )}). In eight strata, all meals and snacks were provided and the almonds were said to have been consumed as a snack (Berryman et al. ⁽ Reference Berryman, West and Fleming ¹¹ ) or incorporated into the meals and snacks (Jia et al. strata 1 and 2⁽ Reference Jia, Li and Zhang ^{15 )}; Li et al. ⁽ Reference Li, Liu and Liu ^{25 )}; Lovejoy et al. strata 1 and 2⁽ Reference Lovejoy, Most and Lefevre ^{26 )}; Sabaté et al. strata 1 and 2⁽ Reference Sabaté, Haddad and Tanzman ^{18 )}). The form of almonds that was used was described only by Berryman et al. ⁽ Reference Berryman, West and Fleming ^{11 )}, who reported administering unsalted, whole, natural almonds with skins, and by Jia et al. strata 1 and 2⁽ Reference Jia, Li and Zhang ^{15 )}, who reported using almond powder. In the remaining five strata in which all meals and snacks were provided (Li et al. ⁽ Reference Li, Liu and Liu ^{25 )}; Lovejoy et al. strata 1 and 2⁽ Reference Lovejoy, Most and Lefevre ^{26 )}; Sabaté et al. strata 1 and 2⁽ Reference Sabaté, Haddad and Tanzman ^{18 )}), it is assumed that whole almonds, almond pieces and ground almonds were used to prepare the meals.

Control foods/diets

Although all studies were randomised and controlled, the control food was not defined in some studies but defined in other studies. In thirteen of the twenty-seven strata, subjects in the control group or during the control phase were instructed not to consume nuts, but were not provided with a control food or with a control diet (Abazarfard et al. ⁽ Reference Abazarfard, Salehi and Keshavarzi ^{9 )}; Sweazea et al. ⁽ Reference Sweazea, Johnston and Ricklefs ^{10 )}; Foster et al. ⁽ Reference Foster, Shantz and Vander Veur ^{16 )}; Kurlandsky & Stote strata 1 and 2⁽ Reference Kurlandsky and Stote ^{14 )}; Ruisinger et al. ⁽ Reference Ruisinger, Gibson and Backes ^{27 )}; Tamizifar et al. ⁽ Reference Tamizifar, Rismankarzadeh and Vosoughi ^{21 )}; Tan & Mattes strata 1 to 4⁽ Reference Tan and Mattes ^{22 )}; Wien et al. ⁽ Reference Wien, Sabaté and Iklé ^{28 )}; Wien et al. ⁽ Reference Wien, Bleich and Raghuwanshi ^{23 )}). In fourteen strata, either a ‘control food’ (e.g. cheese sticks or a muffin or olive oil) was provided (Cohen & Johnston⁽ Reference Cohen and Johnston ^{24 )}; Damasceno et al. ⁽ Reference Damasceno, Perez-Heras and Serra ^{19 )}; Jenkins et al. strata 1 and 2⁽ Reference Jenkins, Kendall and Marchie ^{20 )}; Spiller et al. strata 1 and 2⁽ Reference Spiller, Jenkins and Bosello ^{17 )}), or the entire control diet was provided (Berryman et al. ⁽ Reference Berryman, West and Fleming ^{11 )}; Jia et al. strata 1 and 2⁽ Reference Jia, Li and Zhang ^{15 )}; Li et al. ⁽ Reference Li, Liu and Liu ^{25 )}; Lovejoy et al. strata 1 and 2⁽ Reference Lovejoy, Most and Lefevre ^{26 )}; Sabaté et al. strata 1 and 2⁽ Reference Sabaté, Haddad and Tanzman ^{18 )}).

Study quality

Based on Health Canada's quality appraisal tool, all of the studies were considered to be ‘higher quality’^{( 8 )}. Across all eighteen publications, the most commonly identified limitations included the lack of reporting on allocation concealment (n 16), the lack of reporting on the method of randomisation and thus the ‘appropriateness’ of the randomisation method, which also is a quality factor, could not be determined (n 13), as well as the lack of reporting of an intent-to-treat analysis (n 13).

Effects of almonds on fasting blood lipids

The fasting blood lipids that were assessed in each of the twenty-seven strata, as well as other information pertinent to the subgroup analyses (i.e. study design, almond intake, whether a control food/diet was provided, study duration, and baseline blood lipid levels), are summarised in Table 2. In the study by Lovejoy et al. strata 1 and 2⁽ Reference Lovejoy, Most and Lefevre ^{26 )}, baseline TAG levels were not reported; thus, a determination as to whether TAG levels at baseline were or were not optimal could not be made. In the study by Spiller et al. strata 1 and 2⁽ Reference Spiller, Jenkins and Bosello ^{17 )}, TC, LDL-C, HDL-C and TAG were assessed at baseline and at the end of treatment; however, only the results for TC and LDL-C at the end of treatment were reported. Thus, the HDL-C and TAG results could not be included in the meta-analyses, and in order to include the results related to TC and LDL-C in the meta-analyses, the mean difference in the effect for TC and LDL-C had to be calculated by subtracting the end-of-treatment value in the control group from the end-of-treatment value in the almond group (as opposed to subtracting the changes from baseline in the control group from the changes from baseline in the almond group, which was done for all other parallel studies).

Table 2. Summary of study design and duration, almond dose and baseline blood lipids

TC, total cholesterol; LDL-C; LDL-cholesterol; HDL-C; HDL-cholesterol; P, parallel; X, crossover; Pr, provided; NPr, not provided; BL, baseline; EOT, end of treatment; N.O., not optimal; O, optimal; –, not reported.

* Mean baseline TC, LDL-C, HDL-C and TAG were categorised as O or N.O., based on the targets established in the National Cholesterol Education Program Adult Treatment Panel III guidelines (i.e. optimal blood lipid levels were defined as: TC < 5·17 mmol/l; LDL-C < 2·59 mmol/l; HDL-C ≥ 1·03 mmol/l; TAG < ·69 mmol/l).

† In the study by Spiller et al. strata 1 and 2⁽ Reference Spiller, Jenkins and Bosello ^{17 )}, HDL-C and TAG levels were assessed at BL and at EOT; however, values were presented only in figure form, with no measures of variability. The results related to HDL-C and TAG could not be included in the meta-analyses.

The effects of almonds on fasting TC levels were assessed in all twenty-seven strata. The daily almond intake was 45 g or greater in 63 % of the strata, the design was crossover in 37 % of the strata, the baseline fasting TC level was not optimal in 52 % of the strata, a control food/diet was provided in 52 % of the strata, and the study duration was <12 weeks in 78 % of the strata (Table 3). As can be seen in Table 3 and Fig. 2, the reduction in TC was statistically significant when data from all twenty-seven strata were pooled (−0·153 mmol/l; 95 % CI −0·235, −0·070 mmol/l; P < 0·001). As there was no publication bias identified, no adjustment to these values was made. In all of the subgroup analyses, the pooled effect sizes for TC were negative. Statistical significance was observed when pooling those strata in which the almond dose was ≥45 g/d, the baseline TC level was not optimal, the study design was either parallel or crossover, the control food/diet either was or was not provided, and the duration of the almond intervention period was <12 weeks (Table 3).

Fig. 2. Effect of almond consumption on total cholesterol (TC).

Table 3. Effects of almonds on blood lipid levels: results of meta-analyses of randomised controlled trials* †

TC, total cholesterol; LDL-C, LDL-cholesterol; HDL-C, HDL-cholesterol; NA, not applicable.

* For each meta-analysis, the results indicate (in vertical order from top to bottom): n (the number of strata), the pooled effect/point estimate, the 95 % CI, and the statistical significance of the point estimate. Subgroup analyses were conducted if there were at least three or more strata.

† An assessment of publication bias was conducted for each lipid parameter, but only for the meta-analysis that included all strata. Publication bias was not identified for TC, LDL-C, HDL-C or TAG. For the ratio of TC:HDL-C, two studies were found to be missing to the right of the pooled effect, and with the missing studies imputed, the pooled effect was −0·120 (95 % CI −0·289, 0·050). For the ratio of LDL-C:HDL-C, two studies were found to be missing to the right of the pooled effect, and with the missing studies imputed, the pooled effect was −0·059 (95 % CI −0·175, 0·056).

The effects of almonds on fasting LDL-C were assessed in twenty-five strata. The daily almond intake was 45 g or greater in 60 % of the strata, the design was crossover in 40 % of the strata, the baseline fasting LDL-C level was not optimal in 80 % of the strata, a control food/diet was provided in 48 % of the strata, and the study duration was <12 weeks in 76 % of the strata (Table 3). As can be seen in Table 3 and Fig. 3, the reduction in LDL-C was statistically significant when data from all twenty-five strata were pooled (−0·124 mmol/l; 95 % CI −0·196, −0·051 mmol/l; P = 0·001). As there was no publication bias identified, no adjustment to these values was made. In all of the subgroup analyses, the pooled effect sizes for LDL-C were negative, except for when the five strata in which the baseline blood lipid level was optimal were pooled. For the subgroup analyses, statistically significant pooled reductions in LDL-C were observed when pooling those strata in which the almond dose was ≥45 g/d, the baseline LDL-C level was not optimal, the study design was crossover, a control food/diet was provided, and the study duration was <12 weeks (Table 3). It should be noted that when the control-adjusted changes in LDL-C for the thirteen strata in which the control food/diet was not provided were pooled, the reduction in LDL-C approached statistical significance (P = 0·068).

Fig. 3. Effect of almond consumption on LDL-cholesterol (LDL-C).

Almond consumption was not associated with any significant effect on fasting HDL-C, either in the overall analysis in which all twenty-two strata were pooled or in any of the subgroup analyses (Table 3 and Fig. 4).

Fig. 4. Effect of almond consumption on HDL-cholesterol (HDL-C).

The effects of almonds on fasting TAG were assessed in twenty-five strata. The almond intake was 45 g or greater in 60 % of the strata, the design was crossover in 40 % of the strata, the baseline fasting TAG level was not optimal in 30 % of the strata, a control food/diet was provided in 48 % of the strata, and the study duration was <12 weeks in 76 % of the strata (Table 3). As can be seen in Table 3 and Fig. 5, the reduction in TAG was statistically significant when data from all twenty-five strata were pooled (−0·067 mmol/l; 95 % CI −0·132, −0·002 mmol/l; P = 0·042). As there was no publication bias identified, no adjustment to these values was made. In all of the subgroup analyses, the pooled effect sizes for TAG were negative but not statistically significant, except for when the fifteen strata in which the study design was parallel were pooled, and the resultant pooled effect was a statistically significant reduction in TAG (−0·111 mmol/l; 95 % CI −0·204, −0·017; P = 0·020). Through additional sensitivity analyses, it was determined that this effect was dependent on the inclusion of the parallel study by Abazarfard et al. ⁽ Reference Abazarfard, Salehi and Keshavarzi ^{9 )}, which included 100 females and was a relatively larger study.

Fig. 5. Effect of almond consumption on TAG.

With regards to the ratio of TC:HDL-C, when data from all nine strata were pooled, the reduction in the ratio was statistically significant (see Table 3 and Fig. 6). However, publication bias was detected. Using trim and fill, two studies were found to be missing to the right of the pooled effect size, and with these studies imputed, the pooled effect, though negative (i.e. favourable), was smaller and no longer statistically significant (Table 3). Results for the subgroup analyses were in the same direction of effect (i.e. the pooled effect was negative), with variable statistical significance. With regards to the ratio of LDL-C:HDL-C, when data from all ten strata were pooled, the reduction in the ratio was not significant (−0·089; 95 % CI −0·209, 0·031; P = 0·145) (see Table 3 and Fig. 7). As for the ratio of TC:HDL-C, publication bias was detected, and using trim and fill, two studies were found to be missing to the right of the pooled effect size for LDL-C:HDL-C. With these studies imputed, the pooled effect, though negative (i.e. favourable), was smaller and remained non-statistically significant (Table 3). Results for the subgroup analyses were in the same direction of effect (i.e. the pooled effect was negative), with variable statistical significance.

Fig. 6. Effect of almond consumption on the ratio of total cholesterol:HDL-cholesterol (TC:HDL-C).

Fig. 7. Effect of almond consumption on the ratio of LDL-cholesterol:HDL-cholesterol (LDL-C:HDL-C).