Roselle

Hibiscus Sabdariffa (of the Malvaceae family) is the plant where its leaves and stems are commonly referred to as Hibiscus tea (whereas other species of the Hibiscus family such as Hibiscus Macranthus are not commonly used as teas)^[1] and sometimes is referred to the colloquiol name of Roselle^[2] or Sour Tea.^[3]

The calyx (flowers) of the plant are commonly used for teas, where the white calyx of Hibiscus Sabdariffa has the name 'Isakpa' by the Yorubas of southwestern Nigeria,^[2] 'Krachiap daeng' in Thailand,^[4] 'flor de Jamaica' or simply 'Jamaica' in Mexico,^[5] and Lo-Shen.^[6] It has limited traditional medicinal use, and tends to be a treatment for high blood pressure, gastrointestinal disorders, diaphoresis, and anuria.^[7][4] The red flowers have sometimes been used traditionally for their pigmentation as a dye, which may have a role in histological studies.^[8]

In West African folk medicine, Hibiscus Sabdariffa appears to be recommended as an aphrodisiac.^[9][10]

The following compounds and the quantities are listed below, and those found in the aqueous extracts are most likely found in teas made from Rosella (anthocyanins found in higher amounts in red Roselle):

• Hibiscus Acid^[11][12] and Hibiscus Acid hydroxyethylester^[13]
• The Anthocyanin Delphinidin as the glycosides 3-O-Sambubioside (32.4mg/g aqueous extract)^[14] and 3-Glucoside^[15]
• The Anthocyanin Cyanidin as the glycosides 3-O-Sambubioside (11.5 mg/g aqueous extract)^[14] and 3-Glucoside^[15]
• Quercetin at 11.5 mg/g in the aqueous extract,^[14] richer than a few other herbs;^[16] a 3-sambubioside, 3-glucoside, and 3-rutinoside conjugate are present^[12]
• Kaempferol^[13] as 3-coumaroylglucoside^[12] and 3-Sambubioside (known as Leucoside)^[17]
• Myricetin as 3-arabinogalactose^[12]
• Daidzein (one of the two soy isoflavones)^[16]
• Tiliroside^[11]
• Chlorogenic Acid^[11][13] at 2.7mg/g in the aqueous extract^[14]
• 5-O-Caffeoylshikimic acid, Coumaroylquinic acid,^[17] and Hydroxycitric acid^[12]
• Tetra-O-methyljeediflavanone^[17]
• N-feruloyltyramine^[13] at 0.6% dry weight of the calyx^[12]
• Methyl Digallate^[13]
• Vitamin C^[2]
• (Seeds) Dihydrosterculic (1.57% of the oil, which is 17.35% total seed weight) and Vernolic acid (3.52%), uncommon dietary fatty acids^[18]
• (Seeds) Common dietary fatty acids palmitic (18.52-20%), oleic (18.52-28%), linoleic (40.1-44.72%), and stearic (4.31-5.3%)^[18][19]
• (Seeds) VItamin E, specifically γ-tocopherol^[19]
• (Seeds) β-sitosterol, Daucosterol, Clerosterol, δ-5-avenasterol, and cholesterol^[19]

Although one study noted that water extracts of Roselle appear to be poor sources of phenolics overall, with the red and white variants having 2.3+/-0.8mg/100g and 1.7+/-0.2mg/g respectively and a lesser content of flavonoids (0.1-0.2mg/100g),^[2] another study noted that phenolics and flavonoids reached 58.80+/-1.34mg/g and 13.57+/-0.65mg/g respectively.^[11]

The relative amount of phenolics that are anthocyanins are relatively high, at 0.45% of dry weight of the red calyx.^[12] This is approximately half the concentration of anthocyanins found in blackberry juice, according to one study.^[20]

Anthocyanins may be subject to degradation when exposed to heat, with one study noting an activation energy of 54kJ/mol in the temperature range of 80-90°C^[21] and another noting a range of 47-61kJ/mol.^[20]

Anthocyanins may be degraded by heat treatment, although several human studies still use tea as a vessel for experimenting with Roselle (so it appears that this heat destruction may not be practically relevant)

One study with oral intake of 130.25mg Anthocyanins (65.56mg Cyanidin-3-Sambubioside and 62.12mg Delphinidin-3-Sambubioside) has failed to find appreciable serum levels following ingestion of the tea,^[15] while another study measuring the serum noted a C_max for Delphinidin-3-Sambubioside (1.26ng/mL) and Cyanidin-3-Sambubioside (2.23ng/mL) at the T_max of 90 minutes (both compounds) with AUC values of 2.59ng/mL/h and 4.77ng/mL/h respectively.^[22]

Urinary concentrations of Delphinidin-3-Sambubioside, Cyanidin-3-Sambubioside, or anthocyanin conjugates following oral ingestion of 130.25mg Anthocyanins via Hibiscus Sabdariffa tea (65.56mg Cyanidin-3-Sambubioside, 62.12mg Delphinidin-3-Sambubioside, 0.15mg Cyanidin-3-Glucoside, and 2.42mg Delphinidin-3-Glucoside) have been noted,^[15] and the same research group has conducted another study where 147.4mg Anthocyanins (62.6 Cyanidin-3-Sambubioside and 81.6mg Delphinidin-3-Sambubioside) was given orally to participants noted that detection of these glycosides can be found in the urine within 2 hours, with a maximum excretion rate of 7.51mcg/h; Cyanidin and Delphinidin glucosides could not be determined due to low urinary concentrations.^[22] This latter study noted that the amount of anthocyanins detected in the urine is about 0.018% of the oral dose, with studies on anthocyanins in general suggesting most of the metabolism occurs in the colon by intestinal microflora.^[23]

The anthocyanins have been tested for pharmacokinetics in humans, and they appear to be fairly rapidly excreted with poor bioavailability; the glycosides themselves (Delphinidin-3-Sambubioside rather than Delphinidin) are found in the urine, but the overall exposure to anthocyanins after oral administration is poor

Oral administration of a polyphenolic enriched Hibiscus Sabdariffa extract to rats resulted in a variety of pharmacokinetic profiles for active ingredients of this herb.^[13]

Hibiscus acid at an oral dose of 244.1umol reached a C_max of 112.50+/-4.57uM and took over 2 hours to do so, while its conjugate (Hibiscus Acid hydroxyethylester; 125.3umol) peaked at 6.07+/-0.77uM at a similarly slow time.^[13]

Quercetin (6.9umol) was found to peak at 1.57+/-0.18uM and six glucuronide metabolites (4 diglucuronide and 2 monoglucuronide) were found in plasma.^[13]

N-feruloyltyramine was found to peak after 20 minutes to 0.54+/-0.16uM following oral ingestion of 1uM, and conferred a half-life of 47.8 minutes.^[13]

Some serum parameters are noted with the non-anthocyanin compounds in rats

A study giving 10g of Hibiscus Sabdariffa tea acutely noted urinary concentrations of Cyanidin-3-Sambubioside (18.7+/-11.4mcg) and Delphinidin-3-Sambubioside (6.54+/-3.41mcg) with a metabolite of the latter known as Delphinidin Monoglucuronide (3.92+/-1.60mcg) with total Anthocyanins in the urine measuring at 29.1+/-12.6mcg over 24 hours;^[15] this was following oral ingestion of 130.25mg Anthocyanins with a known Cyanidin-3-Sambubioside (65.56mg) and Delphinidin-3-Sambubioside (62.12mg) content.^[15]

Consumption of Roselle tea appears to be able to reduce diclofenac serum levels (to 62% of the levels achieved normally without Roselle), but also reduced the variability in urinary excretion rates of diclofenac.^[24]

Some minor anxiolytic effects have been noted with the aqueous extract (300mg/kg) and ethanolic (50-300mg/kg) with an ethyl acetate fraction potentiating phenobarbitol-induced sleep time and reducing overall sleep time after repeated dosing.^[25]

Oral ingestion of 25-50mg/kg of Hibiscus Sabdariffa water extract for 28 days in rats increased the activity of Na+/K+ATPase (218-256% of baseline) and Ca²⁺/Mg²⁺ATPase (600-752%) in cardiac tissue;^[26] this may underlie the ability of Hibiscus Sabdariffa to attenuate cardiac contractility in response to adrenaline.^[27] These effects do not appear to be associated with any overt cardiac toxicity after 28 days of ingestion and do not appear to follow dose-response.^[26]

One study with rats using 50-200mg/kg of the calyx water extract for 10 weeks that noted reductions in blood pressure with all doses (not significantly different from each other, occurring at week 4) as well as a prevention of cardiac enlargement (which occurred in control).^[28] This study also noted that the surface area of cardiac capillaries (blood vessels) was increased 59-85.9% in a dose-dependent manner relative to control and the length of capillaries increased 56.9-77.6% (with the middle dose being most effective).^[28] The authors hypothesized that the reduction in cardiac size over time may be related to ACE inhibition, as this has been noted in vitro with Hibiscus Sabdariffa via competitive inhibition^[10] and this phenomena has been noted to occur with pharmaceutical ACE inhibitors.^[29][30]

Mechanistically speaking, a water extract of Hibiscus Sabdariffa is able to concentration dependently relax arteries precontracted by high potassium (27.9% relaxation at 1mg/mL; the EC₅₀ relative to this being 3.37+/-0.26mcg/mL) or phenylephrine (86.01% at 1mg/mL; EC₅₀ of 3.83+/-0.18mcg/mL); it should be noted that these are not EC₅₀ values relative to contraction per se, but relative to maximal contraction.^[31] This relaxation was inhibited by L-NAME and Methylene Blue as well as removal of the endothelium, suggesting they are related to the Nitric Oxide system.^[31]

An ACE inhibitor potential of Roselle has been noted via competitive inhibition,^[10] which is related to the anthocyanins Delphinidin and Cyanidin as 3-Sambubioside glycosides with IC₅₀ values of 141.61μM/84.55μg/mL (K_i of 31.9μM) and 117.75μM/68.41μg/mL (K_i 56.9μM) respectively;^[32] both of these were less effective than the active control, Lisinopril, but greater than that of other bioflavonoids such as apigenin, luteolin, and Quercetin glycosides.^[32] A basic water extract seemed to outperfom the anthocyanin rich fragment in this study (IC₅₀ 40.04μg/mL versus 91.2μg/mL) suggesting other bioactives not in the anthocyanin class are more active.^[32]

Mechanistically, the extract does possess ACE inhibitory potential via the anthocyanins although these are not as strong as pharmaceutical reference drugs.

A study where 10g dried calyx infused in 510mL water (9.6mg anthocyanins) drunk as tea daily before breakfast for 4 weeks was associated with an 11% reduction in systolic blood pressure (139.05 to 123.73mmHg) and 12.5% reduction in diastolic blood pressure (90.81 to 79.52mmHg) in persons with diagnosed hypertension not currently taking other antihypertensive agents.^[5] When compared to the active control of 50mg Captopril (taken in two divided doses) for the same period of time was not significantly different in effects.^[5] These effects are noted in both prehypertensive and hypertensive adults over 6 weeks, but with thrice daily ingestion of 240mL tea with main meals (1.25g of the calyx each serving with 7.04mg Anthocyanins) in a blinded manner (placebo tea sweetened and colored to mimick Roselle tea) noted a reduction of systolic and diastolic blood pressure by 5.5% and 4% respectively in hypertensive persons and to similar level in prehypertensive persons.^[33] Mean Arterial Pressure was reduced significantly by 4.7% when compared to baseline, but barely missed statistical significance relative to placebo (P=0.054).^[33] Although extending for 6 weeks, significant reductions in both blood pressures were noted by week 2,^[33] and other studies using Hibiscus Sabdariffa tea over 15 days have replicated these reductions in blood pressure, with measurements on day 12 reaching reductions of 11.2% systolic and 10.7% diastolic in essential hypertensive persons.^[3] This latter study had reductions in blood pressure about two-fold observed in the previous study at the 2 week interval, but this may simply be due to this demographic having higher blood pressure at the start of the study.^[33][3]

When compared to Lisinopril, the 12.21% blood pressure reduction observed with 250mg Anthocyanins was less than that of 10mg Lisinopril.^[34]

In type II diabetics, Roselle tea consumed over 30 days was able to reduce systolic blood pressure from 134.4+/-11.8 to 112.7+/-5.7 (16.1% reduction) when the control group drinking black tea actually had an increase (118.6+/-14.9 to 127.3+/-8.7; 7.3% increase); Roselle also decreased mean pulse pressure to 66% of baseline values.^[35]

Repeated human trials have noted a reduction in blood pressure associated with drinking moderate amounts of Roselle tea. Only one trial has attempted to blind participants to the treatment by giving them a falsely scented and flavored placebo tea, but this also replicated the benefits associated with Roselle

In hypercholesterolemic persons (220mg/dL or greater), 12 weeks of Roselle extract standardized to either 10mg (capsules) or 20mg (via tea) total Anthocyanins was able to reduce serum triglycerides^[36] and this has been replicated with 100mg of Roselle extract (19mg Anthocyanin Sambubiosides) where over 31 days in persons with metabolic syndrome triglycerides were reduced 23% relative to baseline (no control group) and HDL-C increased 10%; this latter study failed to note any reductions of LDL-C or Total cholesterol.^[37]

When making tea from the calyxes (2g steeped for 25-30 minutes and drunk; twice a day for a month), type II diabetic persons have noted increases in HDL-C (16.7%) and decreases in ApoB100 (3.4%), total cholesterol (7.6%) and LDL-C (8%). The increase in APo-A1 (4.6%) slightly missed being statistically significant.^[38]

In hypercholesterolemics with serum LDL-C in the 130-190mg/dL range who received 1g of the plant extract daily (hydroalcoholic leaf extract) for 90 days did not influence body mass nor did it significantly influence any measured parameter in serum (Total cholesterol, LDL-C, HDL-C, and blood glucose).^[39]

Improvements may occur in regards to cholesterol and lipoproteins in persons with metabolic ailments (currently hypercholesterolemia and diabetes studies in humans), although the degree of improvement is moderate relative to other compounds

In regards to triglycerides, 5-15% of the diet as Hibiscus Sabdariffa ethanolic extract for 4 weeks noted decreases in total lipids (12.3%) and triglycerides (48%),^[40] with another study using 5-10% of the calyx itself reaching 49-53% reduction of serum lipids.^[41] A mechanistic study thought that this may be related to activation of AMPK, which was noted in hepatocytes and suppressed SREBP-1 (a protein mediating hepatic fatty acid synthesis).^[42]

The absorption of triglycerides has been noted to be significantly reduced at 5% of the diet as Hibiscus Sabdariffa in rats (reducing apparent fatty acid absorption from 95.1% to 91.4%), as assessed by fecal analysis; 10-15% were associated with less fecal weight secondary to less food intake.^[43]

In human studies, one study in diabetics notes a 14.9% decrease with 2g of calyx steeped for 25-30 minutes and drunk twice a day for a month^[38] with a study in hypercholesterolemics using 1g of a hydroalcoholic extract over 90 days failing to reach statistical significance.^[39]

Studies in rats suggest potent hypolipidemic (triglyceride lowering) effects of Hibiscus Sabdariffa possible related to hepatic fatty acid synthesis suppression and possible fatty acid burning via AMPK; studies in humans using lower doses (commonly used for blood pressure health) tend to have far less potency

Roselle appears to inhibit the alpha-glucoside enzyme with more potency using Red Hibiscus (IC₅₀ 25.2ug/mL) than the White variant (47.4ug/mL) although the opposite trend exists on the alpha-amylase enzyme with the White variant being more potent (90.5ug/mL) than the red (187.9ug/mL); all four of these values were weaker than the active control of Arcabose.^[2]

In diabetic rats (streptozotocin), 100-200mg/kg Roselle daily was able to reduce the spike in fasting glucose by approximately 60-65% and the both doses significantly reduced the increase in serum insulin; 200mg/kg almost normalizing insulin relative to non-diabetic control.^[11] Similar potency has been noted elsewhere with an alloxan-induced diabetic model, where an ethanolic extract of Hibiscus Sabdariffa at 200mg/kg was associated with preventing 57% of the increase in blood glucose as well as partly attenuating the increases in lipids, cholesterol, and the artherogenic index (the latter by 32%); all of somewhat similar efficacy as 10mg/kg Lovastatin.^[44]

A human study using 100mg of Hibiscus Sabdariffa extract (19% anthocyanin sambubiosides) for one month noted reductions in blood glucose (7% reduction from baseline) which outperformed the diet condition, although pairing diet and the extract improved the blood glucose reduction to 9%.^[37]

In 3T3-LI adipocytes, an inhibition of adipocyte differentiation can be achieved at very high concentrations (500mcg/mL) with an equally weak inhibition of triglyceride uptake (IC₅₀ 799+/-225mcg/mL).^[17] This same study noted that a purified extract of Roselle for polyphenolics only improved the IC₅₀ to 9.1+/-2.8mcg/mL, and when fractionating further these effects were thought to be due to delphinidin and cyanidin-3-sambubiosides, chlorogenic acid, tetra-O-methyljeediflavanone, and other glycosylated flavonoids.^[17]

120mg/kg of a Hibiscus extract (28% anthocyanins) to obese rats for 60 days in two daily doses of 60mg/kg noted that obese mice had body weight gain suppressed with significance at 7 weeks, with normal weight mice noting a trend towards suppressed body weight but not reaching significance.^[45] This study noted decreased food intake, which was thought to underlie the effects observed.

Hibiscus Sabdariffa anthocyanins up to 200mg/kg bodyweight in rats for 5 days has failed to significantly influence estrogen levels in serum, and have failed to alter uterine weight (indicative of estrogenic effects independent of estrogen serum concentrations).^[46]

Hibiscus Sabdariffa anthocyanins up to 200mg/kg bodyweight in rats for 5 days has failed to significantly influence testosterone or testicular weight.^[46]

A study dissolving 10g of Hibiscus Sabdariffa (130.5mg Anthocyanins) in 200mL water for acute consumption with a breadroll which measured serum for the next 10 hours noted increases in serum antioxidant potential (measured by FRAP) despite no changes in uric acid, although an increase in Vitamin C AUC was noted.^[15]

Serum levels of MCP-1 (Monocyte Chemoattractant Protein-1 is a biomarker of inflammation), after consumption of the calyx extract at 10g, are reduced by 17% within 90 minutes; this study attempted to measure IL-6 and IL-8 but no detectable levels were determined at any time in control or with Hibiscus.^[12]

The aqueous extracts of Roselle appear to reduce protein expression of αENaC after oral administration to adrenalectomized rats, which may be related to the diuretic and natiuretic (sodium excreting) effects of Roselle.^[14] This diuretic effect is known as potassium sparing, and works in a dose-dependent manner between doses of 1500-2500mg/kg (authors stating that, due to extraction, this equates to 5mg/kg) although the highest dose was less effective than 13mg/kg furosemide.^[47]

Appears to have diuretic properties

A study using drug-induced calcium kidney stone formation in rats treated with 250-750mg/kg Roselle extract was able to significantly reduce calcium deposition in renal tissue (no dose dependence noted, underperformed relative to 750mg/kg Cystone) as assessed by histology; the increases in phosphorus and urea (in kidney stone control) was reduced with comparable potency to Cystone, the reference drug.^[48]

In persons ingesting 1.5g of the dried calyx as tea twice a day, it appears that in persons who have not had a kidney stone previously that a trend exists towards increased uricosuria (uric acid excretion in the urine) while in those who have formed a kidney stone this relationship is statistically significant;^[49] the effect was noted to be transient, and vanished during a washout period.

May possess kidney stone reducing properties

When fed to rats at an oral dose of 25-50mg/kg (aqueous extract) over 28 days was able to reduce the activity of Ca²⁺/Mg²⁺ATPase (no apparent dose response), without influencing NA²⁺/K⁺ATPase (although a trend to increase); plasma albumin or ALP was noted, but a decrease of urea and creatinine were noted at the higher dose of 50mg/kg.^[6]

In rats given a 5/6 Nephrectomy and then divided into either control or a group treated with Roselle (250mg/kg of a water extract from red calyxes for 6 weeks after a 1 week downtime after surgery) noted significantly reduced glomerular and tubulointerstitial injuries (histological examination) and less fibrosis with no significant effect on albuminuria or serum IL-6/TNF-α levels.^[4] The increase in blood pressure associated with 5/6 Nephrectomies was significantly attenuated.^[4]

Blood pressure related mechanisms have been noted in renal tissue, which may be related to those observed in human trials

The non-anthocyanin portions also have hepatoprotective effects, as evidence by a pigment free extract (with a high concentration of green tea catechins including EGCG at 20%) able to protect from acetominophen induced hepatotoxicity.^[50]

A study in persons with metabolic syndrome noted that 31 days of 100mg Roselle extract (19mg Anthocyanin Sambubiosides) was able to reduce serum ALT and AST, suggesting Roselle attenuates hepatic damage.^[37]

One study investigating rat testes noted that four weeks of 200mg/kg Hibiscus extract in male mice (a dose similar to those used in other studies) was able to adversely affect sperm morphology.^[51] The percentage of abnormal sperm morphology (nonspecific morphology abnormalities) increased from 18.5% at baseline to 43.5-52.5%; sertoli cells also appeared to be altered. Another study over the course of 5 days failed to find any abnormalities with the testes beyond a slight decrease in testicular protein content with 100mg/kg (none at 200mg/kg)^[46] and a third notes that large doses between 1.6-4.2g/kg of the water extract did not alter testicular body weight ratio, with the two higher doses reducing sperm count and altering sperm and tubular morphology.^[9]

Although a possible estrogenic effects is though to exist, a failure to alter testicle weight was noted at doses even up to 4.2g/kg bodyweight for 12 weeks^[9] and a reduction of the testicular:body weight ratio tends to be indicative of estrogen-related testicular toxicity.^[10] Currently the mechanism mediating possible testicular toxicity are unknown, with the authors of the aforementioned rat study citing the following and hypothesizing Quercetin may be related, as Quercetin isolated from Roselle coloring was noted to possess mutagenic activities in vitro.^[52]

There appears to be some possible testicular toxic effects, seen at the dose of 200mg/kg (human dose of about 2g for a 150lb person) or above. The mechanisms underlying this possible toxicity are not known, and no studies exist in humans assessing testicular toxicity

Morus Alba (White Mulberry) is a herb where the leaves and stems are commonly drunk as tea for anti-diabetic purposes, and this manner of consumption is known to isolate alpha-glucoside enzyme inhibitors and possibly prevent the absorption of carbohydrates. The potency on the alpha-amylase enzyme (that mediates starch digestion) from Morus Alba is relatively weak with 2mg/mL inhibiting 1.17+/-0.74% of the enzyme's action.^[53] Roselle can inhibit the enzyme by 18.99+/-1.39% at this concentration, but adding both together synergistically increases inhibition to 65.75+/-0.60%.^[53]

Chrysanthemum Indicus and Roselle appear to be synergistic in inhibiting the alpha-amylase enzyme in vitro.^[53]

Roselle and Aegle Marmelos appear to synergistially inhibit the alpha-amylase enzyme when incubated at 2mg/mL each.^[53]

One human study using Roselle tea prepared twice a day for 15 days in persons with hypertension specifically investigating renal toxicity failed to find any evidence of toxicity relative to either baseline or control (black tea).^[54]

Looking at rats, a dose-escalation study noted that no mortality existed for acute dosing up until 2,010mg/kg, where 2,020mg/kg was associated with one death and 2,050mg/kg killed all five rats given the dose; the authors suggested that 2g/kg was approximately the maximum tolerated dose,^[48] although a 90 day trial using this dose noted some mortality associated with diarhhea on day 8 with water and ethanolic extracts, with those given a 50:50 extract lasting to day 28; 300mg/kg was not associated with as much death, with sporadic toxic effects on RBCs, increased AST, and food intake^[55] with 120mg/kg over 60 days not exerting any overt toxic effects.^[45]

Despite the aforementioned, a review of safety^[10] has noted that the LD₅₀ to be above 5,000mg/kg citing a rat intervention in hypertensive rats.^[56] The LD₅₀ may be variable due to processing techniques and varying levels of bioactives, as the bioactive(s) underlying this toxicity is currently not known.

2,000mg may be approximately the lowest toxicity threshold, with no acute but possible chronic adverse effects at this dose and higher doses being more likely to induce toxic effects

Due to the potential toxicity, 200mg/kg as a rat dose (32mg/kg human dose, 2.2g of calyx per day for a 150lb person) may be a practical upper limit for usage

Testicular Toxicity falls into this category somewhat (in this particular case, sperm cells are implicated) and discussed in the section of 'Interactions with Organs', subset 'Testicles'

In a study on mice using 0.6g/100mL or 1.8g/100mL of Hibiscus Sabdariffa extract in the drinking water during 21 days of lactation reported suppressed food and water intake in mothers (no dose dependence on the former) and both increased body weight and delayed onset of puberty in rat pups.^[57] This study itself attributed a reduced food intake during lactation to these effects, and a followup study by the same researchers noted a dose-dependent increase in serum sodium and corticosterone, and was again attributed to maternal malnutrition.^[58] These results have been replicated twice again by the same research group,^[59] with one noting that these trials have not noted birth defects per se.^[60]

One trial has noted that the offspring consuming the same doses of Hibiscus Sabdariffa themselves had delayed pubertal onset, again attributed to reduced food and water intake.^[61]

A series of studies that note increased rat pup weight with a possible delay in puberty, which have been linked to maternal malnutrition (no pair fed studies as of yet, which are studies that restrict nutrition in one group to match that of the drug group; attempting to delineate effects of caloric deprivation and effects of the drug. Due to the lack of pair fed studies, this hypothesis is still tentative)