Gentamicin

Limits on what this shows

• Null and contradicting findings: Serum magnesium, calcium, sodium and potassium unchanged in the single-dose adult study [5] and the neonatal study [6]. Dawson 1986 [10] (thesis): 28 gentamicin-treated patients (courses 2 to 17 days, mean 6) against 21 surgical controls without aminoglycosides: "no evidence was found for an effect of gentamicin on electrolyte excretion"; hypomagnesemia and hypocalcemia were very mild and occurred in only a few patients; the author attributes this to short courses. Numeric results not retrieved. In a randomized trial of 60 patients on gentamicin for 7 days or more, no difference in hypokalemia, hypomagnesemia, or potassium and magnesium wasting between the pentoxifylline and placebo arms [41] (not an unexposed comparison; counts not retrieved). A dog study found hypokalemia and hypocalcemia but no hypomagnesemia [57] (listing summary).
• Confounding by indication: Presumed. Critical illness, malnutrition or nil-by-mouth status, diarrhea, diuretics, amphotericin and cisplatin independently lower magnesium, and hypomagnesemia is common in hospital cohorts before any aminoglycoside [20][23]. Only the within-subject changes (a)(b) and the controlled cohort [10] carry attribution.
• At-risk subgroups: Prolonged or high cumulative dosing [12][17], cystic fibrosis with repeated courses (case reports, class data [24]), cancer chemotherapy patients [11], poor oral intake, concurrent diuretics, neonates and preterm infants [6].

Regulatory status

Label: paresthesias, tetany, positive Chvostek and Trousseau signs and confusion have occurred in patients with hypomagnesemia, hypocalcemia and hypokalemia; decreased serum magnesium is a listed laboratory abnormality; adults and infants have needed corrective electrolyte therapy; correct electrolytes as needed [1][2]. Read through summarizing tools, so wording is paraphrased.

Population evidence (Tier A, gentamicin in cohort)

(a) Elliott 2000 [5]: 5 healthy volunteers, gentamicin 5 mg/kg IV, peak serum level 12.8 to 20.6 mcg/mL. Fractional magnesium excretion 3.4% to 11.8% (P=0.03); fractional calcium excretion 1.8% to 6.8% (P<0.01); serum electrolytes unchanged; effect transient; the authors suggest the distal convoluted tubule. Abstract values via a PubMed listing summary; no CI in the abstract; full text not opened. (b) Giapros 2004 [6]: 23 preterm and term neonates treated for suspected infection; serum and 3-hour urine electrolytes before and right after each infusion on days 1, 3, 4 and 7. Fractional excretion of sodium and magnesium and the urinary calcium-to-creatinine ratio rose significantly after infusion in both preterm and term infants; fractional potassium and phosphate excretion did not change; serum electrolytes did not change. Dose and numeric values are not in the abstract. These are independent groups (adult volunteers and neonates) with the same direction. (c) Keating 1977 [11], mixed aminoglycosides: 17 cancer patients developed hypocalcemia, hypomagnesemia and hypokalemia 2 to 8 weeks after gentamicin, tobramycin, amikacin or sisomicin (gentamicin not separated), with magnesium and potassium wasting in 6 and low PTH; 12 died before recovery and doxorubicin may have contributed. Listing summary only.

Case-level evidence (Tier B, gentamicin)

Chou 2005 [12]: 4 normotensive women, cumulative gentamicin 1.2 to 2.6 g, serum magnesium 0.9 to 1.2 mg/dL, fractional magnesium excretion 3% to 6%, hypokalemia 1.8 to 2.3 mmol/L, symptoms lasting 2 to 6 weeks after stopping. Kes and Reiner 1990 [13]: 7 patients with inappropriate urinary losses, magnesium wasting correlated with cumulative dose (abstract). Bar 1975 [14] (2 patients), Patel 1979 [15] (1 patient), Landau 1997 [16] (4 children), Kelnar [17] (a 12-year-old after a cumulative 14,400 mg over four months, aldosterone not raised). Systematic reviews of case reports: Scoglio 2021 [18], 82 patients with aminoglycoside or colistin (the corpus also included capreomycin and viomycin), hypomagnesemia, hypokalemia and hypocalcemia each in more than three quarters of cases, renal magnesium wasting the most common disorder, no prevalence derivable, resolution within 4 weeks in 32 of the 54 cases with timing data (my count from the percentages given); Priyadarshi 2025 [19], 43 antibiotic-induced Bartter-like cases, gentamicin 41.8%, median onset 10 days (IQR 7 to 18), median resolution 14 days (IQR 7 to 31) after stopping. Neither review stratifies electrolyte results by drug.

CLASS EXTRAPOLATION (Tier C, capped at yellow)

Gentamicin is not confirmed in these cohorts. Zaloga 1984 [21]: aminoglycosides cause hypomagnesemia in more than one-third of patients (abstract; no n, no drug name, no comparator). Chernow 1989 [20]: 193 postoperative ICU patients, 61% hypomagnesemic at admission; severe hypomagnesemia (17 patients) more common after aminoglycosides (P<0.001), aminoglycoside unspecified. Deshmukh 2000 [22]: 80 paediatric ICU children and 25 ward controls; 70% hypomagnesemic, more frequent with aminoglycosides and diuretics. Liamis 2012 [23]: 2,284 internal-medicine adults, 4.7% hypomagnesemic, causes mainly diuretics and aminoglycosides. Whang 1984 [56], a narrative review, lists gentamicin as a cause of enhanced magnesium loss (listing summary). Amikacin only: von Vigier 2000 [24] (24 children with cystic fibrosis, amikacin plus ceftazidime for 14 days; plasma magnesium median 0.77 to 0.73 mmol/L, fractional excretion 0.0568 to 0.0721; potassium and calcium unchanged) and Wu 1996 [25] (18 children, amikacin, 50% hypomagnesemic, urinary magnesium-to-creatinine 0.46 to 0.82 mmol/mmol, P<0.05, serum unchanged overall). Cohort test: gentamicin named in none of [20][21][22][23] abstracts; absent from [24][25]. These support plausibility for the class, not a gentamicin effect size.

Mechanism

Present and partly named. Gentamicin raised cytosolic calcium and ERK activity in opossum proximal tubule cells expressing the calcium-sensing receptor [29]; reduced hormone-stimulated magnesium uptake and cAMP in mouse distal convoluted tubule cells [30] (abstract partly truncated); rapidly raised calcium and magnesium excretion in anesthetized rats without a change in GFR [27], and the effect was not altered by parathyroidectomy and resembled furosemide [26]; distal transporters were upregulated within 6 hours, limiting further loss over 7 days in rats [28]. The proposed human site differs between authors (distal convoluted tubule [5]; thick ascending limb [12][26]). The Bartter-like phenotype with low PTH is consistent with the receptor mechanism [12]. An alternative proposed in 1984, drug-induced hyperaldosteronism [58], is contradicted by a case with normal aldosterone [17].

Why orange, not red

The two gentamicin cohorts measured urinary handling with unchanged serum values; the circulating-deficit evidence is case-level or class-level; a controlled short-course cohort found nothing [10]. Boundary: functional loss within hours of dosing; clinically significant circulating deficit mainly after prolonged or heavy exposure. Could rise with full text of [21] (if gentamicin is named) or a controlled gentamicin cohort with serum magnesium.

Counter-evidence held on record

Counter-evidence Serum values stayed normal in both gentamicin cohorts with a measured urinary effect [5][6]; the one controlled cohort found nothing over short courses [10]; the large hospital cohorts are confounded and do not name gentamicin [20][21][22][23]. Framing caution This describes renal loss of magnesium during an injected antibiotic course, mostly in hospital settings. It is a monitoring prompt, not a finding that dietary magnesium is low and not a supplement recommendation. Any replacement in documented hypomagnesemia is for the treating clinician.

Limits on what this shows

• Null findings: Serum calcium unchanged in short-course human studies [5][6]; controlled short-course cohort found no electrolyte effect [10]; amikacin cohort: serum and urinary calcium unchanged over 14 days [24] (class). Neonatal hypocalcemia also occurs without any antibiotic, and [7] has no unexposed comparator, so 15% cannot be attributed to the drug.
• Confounding by indication: Presumed. Late-preterm and sick term neonates have a physiologic calcium nadir, sepsis and prematurity themselves lower calcium, and boys and infants under 37 weeks were over-represented among low values for reasons unrelated to gentamicin [7]. Attribution rests on the within-subject urinary studies (c)(d) and the PTH pattern in cases.
• At-risk subgroups: Neonates on 4 or more days of gentamicin, especially late-preterm and boys [7]; prolonged or high cumulative adult dosing [12]; patients with concurrent magnesium depletion.

Regulatory status

Label: hypocalcemia named with tetany; decreased serum calcium a listed laboratory abnormality; calcium salts may reverse gentamicin-related neuromuscular blockade; tetany and muscle weakness in infants [1][2].

Population evidence (Tier A, gentamicin in cohort)

(a) Chiruvolu 2008 [7]: retrospective analysis of serum calcium in 1,624 neonates of 35 weeks or more given gentamicin once daily for 4 or more days (calcium measured at 47 hours or later). 241 (15%) had calcium below 8 mg/dL; 16.4% of boys versus 11.8% of girls (P=0.01); 23.9% under 37 weeks versus 14.1% at 37 weeks or more (P=0.01). A second value was available in 883; 23.2% of those initially low were still low. No unexposed comparison group. (b) Jackson 2003 [8]: hypocalcemia in term and near-term neonates on gentamicin for 4 or more days after the interval changed from every 12 to every 24 hours; numbers not available from the abstract page. (c) Tan 2015 [9]: 28 full-term newborns on IV gentamicin every 36 hours, pretest-posttest; median urinary calcium-to-creatinine ratio 0.021 (range 0.004 to 0.071) before, 0.043 after the first dose, 0.144 (0.015 to 1.160) after the second dose; described as significant, no P value in the abstract. (d) Fractional calcium excretion 1.8% to 6.8% (P<0.01) in healthy adults [5] and a higher calcium-to-creatinine ratio in 23 neonates [6]. Serum calcium was unchanged in [5] and [6].

Case-level evidence (Tier B)

Hypocalcemia with low or inappropriately low PTH during gentamicin courses: Bar 1975 [14], Patel 1979 [15], Chou 2005 [12] (low PTH persisted with hypocalcemia), Keating 1977 [11] (hypoparathyroidism without parathyroid hyperplasia at autopsy in a mixed-aminoglycoside series).

Mechanism and companion marker

Present: calcium-sensing receptor activation by gentamicin [29] with renal calcium wasting [27], and magnesium depletion, which one case report proposed as a cause of functional hypoparathyroidism [59]. The companion marker moves as expected for the receptor and magnesium mechanism (PTH low, not high) [12][14][15], which also argues against a bone-mobilization or vitamin D explanation (Hard Rule 3). In rats the urinary calcium rise was reported to need an intact parathyroid in one study [31] and not in another [26]; the discrepancy is unresolved.

Why orange, not red

No comparator in the large neonatal cohort, no numbers for [8], serum unchanged in short-course studies, and the adult hypocalcemia evidence is case-level. Boundary: neonates with 4 or more days of exposure; prolonged or heavy adult exposure. Could rise with a neonatal cohort giving calcium in unexposed matched infants.

Counter-evidence held on record

Counter-evidence Baseline neonatal hypocalcemia is common and unadjusted for indication [7]; the abstract-level adult data are small or case-level. Framing caution A monitoring prompt about calcium handling during gentamicin therapy, particularly in newborns. It is not evidence of low calcium intake or of bone loss and does not support calcium or vitamin D supplementation by the reader.

Limits on what this shows

• Null findings: [5][6][24][26] above; no difference in hypokalemia between arms of the pentoxifylline trial [41] (counts not retrieved). No cohort comparing hypokalemia incidence between gentamicin and an unexposed or active comparator was found.
• Confounding by indication: Presumed. Diarrhea, vomiting, diuretics, amphotericin and hyperaldosteronism from illness lower potassium in the same patients.
• At-risk subgroups: Prolonged or high cumulative dosing, concurrent magnesium depletion and diuretic use [12][15][17].

Regulatory status

Label: hypokalemia named alongside hypomagnesemia and hypocalcemia in the tetany and paresthesia warning; decreased serum potassium a listed laboratory abnormality; corrective electrolyte therapy has been required [1][2]. This label statement is the only Tier A anchor for a measured deficit.

Population evidence

Tier A cohorts (no deficit in short courses): fractional potassium excretion and serum electrolytes unchanged across infusions in 23 neonates [6]; serum electrolytes unchanged after a single 5 mg/kg dose in 5 healthy adults [5]. Tier B case series (gentamicin): Chou 2005 [12] serum potassium 1.8 to 2.3 mmol/L in 4 women; Patel 1979 [15] renal magnesium and potassium wasting with raised renin and aldosterone; Kelnar [17] hypokalemia without raised aldosterone; Landau 1997 [16] 4 children with hypokalemic alkalosis; Kes and Reiner [13]; Bar [14]; Shetty 2000 and Alexandridis 2003 [55] (hypokalemic metabolic alkalosis with hypomagnesemia; listing summaries). Systematic reviews of case reports: hypokalemia was present in all 43 Bartter-like cases [19] and in more than three quarters of 82 tubulopathy cases [18], without per-drug stratification. Class (Tier C): Chernow [20] severe hypomagnesemia accompanied by more hypokalemia (P<0.02); amikacin cohort [24] potassium unchanged.

Mechanism

Present: a salt-losing tubulopathy with secondary hyperaldosteronism described in cases [15][19], distal tubule transport effects [5][28], a possible contribution from magnesium depletion (not tested in any opened source). Rats showed unchanged potassium excretion during acute gentamicin infusion [26].

Why orange, not red

One Tier A anchor (the label) and case series; no gentamicin cohort measured a potassium deficit. Boundary: Bartter-like presentations after prolonged or high cumulative exposure. If the label statement is judged a non-cohort source, this entry would be yellow. Could rise with a controlled cohort reporting serum potassium over a full course.

Counter-evidence held on record

Counter-evidence Short-course human studies show no potassium effect [5][6]; the hypokalemia evidence is case-level and shares the same patients and syndrome as the magnesium and calcium entries (see Overlap map). Framing caution A monitoring prompt about potassium loss through the kidney during prolonged injected therapy. It does not show that dietary potassium is low and is not a recommendation to take potassium.

Evidence (Tier B)

Ali 1996 [46] (listing summary): increased urinary copper and zinc excretion in gentamicin-treated rats, more severe in older rats; no values or sample size retrieved. No human zinc study with gentamicin was found.

Why yellow

Mechanistic plausibility from one abstract-level animal observation, probably secondary to tubular injury. Shares its source with the copper entry.

Counter-evidence held on record

Framing caution Not a finding of low zinc in patients and not a basis for zinc supplementation.

Regulatory status

Label: Fanconi-like syndrome in some adults and infants; decreased serum sodium listed [1][2].

Evidence

Giapros 2004 [6]: fractional sodium excretion rose after infusion; fractional phosphate excretion unchanged; serum unchanged. Scoglio 2021 [18]: hyponatremia and hypophosphatemia less frequent than the magnesium, potassium and calcium disorders; proximal dysfunction 3 of 82 (3.7%) and mixed 5 (6.1%). Alexandridis 2003 [55]: a Fanconi-like case with prolonged aminoglycoside use (listing summary).

Hard Rule 2

A tubulopathy is a toxicity outcome. Sodium changes in sick patients are also dilutional or disease-related (Hard Rule 5): no intake implication.

Why yellow

Label plus small cohort null for phosphate; sodium effect measured only as urinary excretion.

Counter-evidence held on record

Framing caution A renal-toxicity monitoring item, not a mineral deficiency and not a reason to change salt or phosphate intake.

Limits on what this shows

• Null findings: The PubMed listing "gentamicin prothrombin time vitamin K coagulopathy" returned 3 records: the two rat papers and a rifampin case report in which gentamicin was given only earlier. No human gentamicin vitamin K study was found.

Regulatory status

No vitamin K statement in the label text retrieved [1].

Evidence (Tier B)

Mi 2016 [52] and Mi 2017 [53] (PubMed listing summaries): rats on a vitamin K-deficient diet with intragastric gentamicin, compared at 14, 21 and 28 days for vitamin K levels and coagulation markers; vitamin K1 restored levels in the second paper. The summaries do not show a gentamicin-only arm, doses or numbers.

Why yellow

Tier B model that uses the drug, but the oral route, deficient diet co-exposure and absence of human data limit relevance to injected gentamicin. Shares the microbiome mechanism hypothesis with the microbiome entry.

Counter-evidence held on record

Framing caution An animal-model observation. It is not evidence that injected gentamicin lowers vitamin K in people and supports no vitamin K supplement claim.

Regulatory status

Boxed warning: potentially nephrotoxic (higher risk with renal impairment, high dose, prolonged therapy); vestibular and auditory toxicity, usually irreversible; numbness, tingling, twitching and convulsions; keep peaks under 12 mcg/mL and troughs under 2 mcg/mL [1].

Evidence

A 1981 narrative review gives nephrotoxicity in 10% to 20% of patients and high-frequency auditory toxicity in about 10% for aminoglycosides in general (listing summary) [54]. In [41] the nephrotoxicity odds ratio was 19.6 (95% CI 3.08 to 114.32) for placebo versus pentoxifylline. The MT-RNR1 warning is under the mitochondrial burden entry [1][35].

Why this is an entry

Included as a context flag so readers do not mistake toxicity outcomes for nutrient findings. No nutrient is measured.

Counter-evidence held on record

Framing caution Hearing loss and kidney injury are drug adverse effects to be monitored by the prescriber. They are not deficiency states and no supplement is implied.

Regulatory status

Label: calcium salts may reverse neuromuscular blockade; electrolytes to be corrected as needed [1]. Patient information page: hypocalcemia, hypokalemia and hypomagnesemia "should be corrected before receiving this medicine" and a caution about herbal or vitamin supplements unless a doctor approves [49] (Tier D patient information, used only to show the framing).

Evidence

Label only for the neuromuscular point. Rat data show a nutrient effect on toxicity in both directions: ascorbic acid reduced injury at 100 mg/kg and worsened it at 200 mg/kg; iron worsened it [42]; copper depletion had no effect [45] (Tier B, listing summaries).

Direction check (Hard Rule 4)

Nutrient affects drug or drug toxicity. Filed as a timing and safety context item with no depletion claim. Intravenous magnesium or calcium given for treatment is a clinician decision; no dose appears in this dossier.

Counter-evidence held on record

Framing caution Not a recommendation to take calcium, magnesium, potassium or antioxidants with gentamicin. Any correction of electrolytes is managed by the treating clinician.

Limits on what this shows

• Why no grade: Dual condition met: no Tier B or Tier C source supports even a yellow grade, and no circulating consumer claim for vitamin D and gentamicin was found. The low PTH seen with hypocalcemia [12][14][15] argues against a vitamin D deficiency mechanism but is not a vitamin D measurement. A disclosed gap, not a finding of safety. Probably a genuine literature gap with a partial tooling limit; re-check with 1,25-dihydroxyvitamin D and neonatal phrasings when access is unthrottled.

What was searched

PubMed listing "gentamicin 25-hydroxyvitamin D" (1 record [50]: a 2018 medical-hypothesis paper on vitamin D in Meniere disease that mentions intratympanic gentamicin; no vitamin D measurement after gentamicin). PubMed listing "aminoglycoside vitamin D calcitriol renal" (10 of 21 records viewed; none studied gentamicin or any aminoglycoside; paricalcitol, calcitriol and puromycin or streptozotocin models). The label text retrieved has no vitamin D statement. A further phrasing ("aminoglycoside vitamin D metabolism gentamicin hydroxylase") was rate limited (HTTP 429) and not retried.

Channel outcomes

PubMed listing pages: two phrasings opened, one rate limited. WebSearch: budget exhausted partway through the review. Europe PMC, Semantic Scholar and Crossref were not usable (network policy or rate limit). Mixed result: two clean zero-signal searches plus tooling limits on a third phrasing.