No ingestion, no inhalation, still toxic: acetone poisoning via skin absorption

Abstract

Background

Acetone toxicity is typically associated with ingestion or inhalation, whereas clinically meaningful transdermal absorption is uncommon and may be overlooked, particularly in patients with impaired skin barriers.

Case presentation

A 48-year-old man with C4 quadriplegia, a chronic sacral ulcer, and cranial cortical stimulation electrodes from the ReHAB trial presented with two weeks of poor oral intake and progressive encephalopathy culminating in respiratory failure. Initial testing showed high anion-gap metabolic acidosis with an elevated osmolar gap (pH 7.22; HCO₃ 6 mmol/L; anion gap 28 mmol/L; measured osmolality 453 mOsm/kg), beta-hydroxybutyrate 7.2 mmol/L, and serum acetone 29.8 mg/dL, with undetectable isopropanol and other toxic alcohols. Additional history revealed topical isopropyl alcohol application to scalp connector sites approximately 36 h before presentation. The patient improved with intravenous fluids and bicarbonate therapy, required brief mechanical ventilation, and returned to baseline mentation by hospital day 9.

Conclusion

This case demonstrates two concurrent mechanisms: starvation ketoacidosis as the primary driver of anion-gap acidosis and transdermal isopropyl alcohol exposure with hepatic conversion to acetone as a plausible contributor to acetonemia and the elevated osmolar gap. Clinicians should consider noningestional exposures in unexplained dual-gap acidosis, especially in patients with neurodisability and device-adjacent or compromised skin.

Clinical trial number

Not applicable

Introduction

Acetone is a volatile organic compound produced endogenously during ketone metabolism and exogenously through metabolism of isopropyl alcohol (IPA) [1, 2]. Clinically significant isopropanol or acetone exposure can cause central nervous system depression, ketonemia, ketonuria, and an elevated osmolar gap; when concomitant ketoacidosis is present, patients may also present with high anion-gap metabolic acidosis [1, 3].

Most reported cases of acetone toxicity are due to ingestion or inhalation. However, systemic toxicity from cutaneous exposure is rarely recognized, particularly in patients with impaired skin barriers or neurologic deficits. We describe a case of acetonemia in a quadriplegic patient after topical isopropyl alcohol was used to clean cranial connector sites, emphasizing the importance of considering transdermal absorption in the differential diagnosis of unexplained encephalopathy.

Case presentation

A 48-year-old man with a complex medical history presented to the emergency department (ED) with acute worsening confusion and altered mental status. His history was significant for a C4 spinal cord injury with quadriplegia following a motor vehicle accident approximately two years earlier; enrollment in the ReHAB (Reconnecting the Hand and Arm to the Brain) clinical trial, involving implantation of eight left-sided cortical stimulation electrodes with intracranial leads and extracranial scalp connector sites for neuromodulation of right upper-extremity function; neurogenic bladder managed with a suprapubic catheter; prior deep vein thrombosis and pulmonary embolism on rivaroxaban; COVID-19-induced cardiomyopathy with recovery of ejection fraction to 60%; chronic hypotension with dysautonomia; ulcerative colitis; obstructive sleep apnea; and a stage IV sacral ulcer status post debridement three weeks before presentation, with prior cultures growing Pseudomonas aeruginosa and methicillin-resistant Staphylococcus aureus.

According to the patient’s significant other, he had not been acting like himself for the preceding two weeks, during which he developed slurred speech, nausea, intermittent choking episodes, poor oral intake, and fluctuating mental status. During the 24 h before presentation, his confusion worsened markedly, prompting emergency medical services activation.

On arrival to the ED, he was hypotensive (87/58 mmHg), tachycardic (110 beats/min), and tachypneic (28 breaths/min). Initial laboratory studies are summarized in Table 1 and were notable for marked leukocytosis (white blood cell count 32.7 × 10⁹/L), anemia (hemoglobin 9.7 g/dL), severe metabolic acidosis (serum total CO₂ 7 mmol/L), and an elevated anion gap of 28 mmol/L. Non-contrast head CT showed no acute intracranial abnormality (Fig. 1). CT of the abdomen and pelvis demonstrated chronic-appearing sclerotic changes consistent with osteomyelitis, without acute fracture. He received 2 L of lactated Ringer’s solution and was admitted to the intensive care unit for management of shock and severe metabolic derangements.

Table 1.

Initial laboratory findings

Laboratory test	Result	Reference range
WBC	32.7 ×10⁹/L	4.0-11.0 ×10⁹/L
Hemoglobin	9.7 g/dL	13.5-17.5 g/dL
Sodium	133 mmol/L	135-145 mmol/L
Potassium	3.3 mmol/L	3.5-5.0 mmol/L
Chloride	98 mmol/L	98-107 mmol/L
Bicarbonate (total CO₂)	7 mmol/L	22-29 mmol/L
Glucose	140 mg/dL	70-140 mg/dL
BUN	3 mg/dL	7-20 mg/dL
Creatinine	0.45 mg/dL	0.7-1.3 mg/dL
Albumin	2.3 g/dL	3.5-5.0 g/dL
Phosphate	1.8 mg/dL	2.5-4.5 mg/dL
Magnesium	1.7 mg/dL	1.7-2.2 mg/dL
Lactate	1.2 mmol/L	0.5-2.2 mmol/L
Beta-hydroxybutyrate	7.2 mmol/L	<0.6 mmol/L
Measured serum osmolality	453 mOsm/kg	275-295 mOsm/kg
Serum acetone	29.8 mg/dL	0-2 mg/dL
Hemoglobin A1c	4.9%	4.0-5.6%

Fig. 1.

Non-contrast head CT demonstrating left-sided cortical electrode implantation

Arterial blood gas revealed pH 7.22, pCO₂ 19.3 mmHg, pO₂ 87 mmHg, and bicarbonate 6 mmol/L, consistent with high anion-gap metabolic acidosis. Further evaluation demonstrated lactate 1.2 mmol/L, beta-hydroxybutyrate 7.2 mmol/L, negative acetaminophen and salicylate levels, and undetectable ethanol. Hemoglobin A1c was 4.9%, and urinalysis was positive for ketones. Calculated serum osmolality was 284 mOsm/kg (based on sodium, glucose, and BUN), while measured serum osmolality was 453 mOsm/kg, yielding an osmolar gap of 169 mOsm/kg. A volatile alcohol screen was sent, and the patient was empirically started on an infusion of D5W 1 L with KCl 40 mEq and NaHCO₃ 150 mEq (8.4%).

The toxicology screen returned positive for acetone at 29.8 mg/dL. Methanol, ethanol, isopropanol, and ethylene glycol levels were all below 1.0 mg/dL. The results were reviewed with the regional poison control center, which recommended supportive care and intravenous fluids. Despite initial resuscitation, the patient became increasingly somnolent over the next 12 h and required endotracheal intubation and mechanical ventilation. At 24 h, arterial pH improved to 7.28 and the anion gap decreased to 21; however, bicarbonate remained critically low at 8 mmol/L. Metabolic acidosis resolved over the next 48 h with continued bicarbonate infusion. Serial acetone levels at 24 and 72 h were 25.6 mg/dL and 15.6 mg/dL, respectively, demonstrating a downward trend that correlated with clinical improvement. The patient was extubated on hospital day 4 and gradually returned to baseline mentation by hospital day 9.

A comprehensive sepsis evaluation, including lumbar puncture, blood cultures, urine culture, and sputum cultures, revealed no clear infectious source. One of two blood culture sets grew Staphylococcus epidermidis and Acinetobacter baumannii, which were considered contaminants. Neurology and infectious diseases consultants concluded that central nervous system infection and electrode-associated infection were unlikely, given the negative cerebrospinal fluid findings.

Further history from the patient’s family clarified that he had not ingested alcohols or toxins. However, approximately 36 h before admission, isopropyl alcohol 70% had been applied topically as both poured solution and wipes to clean the scalp connector sites and the hand interface site used for the ReHAB trial; no occlusive dressing was used. In the setting of two weeks of poor oral intake with ketosis and subsequent acetonemia, cutaneous exposure to isopropyl alcohol with hepatic conversion to acetone was considered a plausible contributor to systemic acetonemia, whereas ketosis was considered the primary driver of the anion-gap metabolic acidosis.

Discussion

Acetone is an organic solvent and a byproduct of both endogenous and exogenous metabolism. It is produced endogenously through spontaneous decarboxylation of acetoacetate and is also the hepatic metabolite of IPA oxidation via alcohol dehydrogenase [1, 2]. Clinically significant acetonemia can be associated with central nervous system depression and an elevated osmolar gap in susceptible patients [1, 3]. Importantly, acetone is an uncharged molecule and therefore does not directly generate an anion-gap metabolic acidosis; when an anion gap is present, it is typically driven by concomitant acid accumulation, such as ketoacids including beta-hydroxybutyrate.

In this patient, the combination of high anion-gap metabolic acidosis and a markedly elevated osmolar gap raised suspicion for toxic alcohol exposure. The toxicology screen was positive for acetone (29.8 mg/dL) and negative for IPA, methanol, ethanol, and ethylene glycol. Although IPA is metabolized to acetone in the liver, the absence of detectable IPA despite exposure approximately 36 h earlier is pharmacokinetically plausible. Isopropyl alcohol has a reported half-life of approximately 2.5 to 8 h, whereas acetone persists longer, with reported elimination half-lives of approximately 17 to 27 h [1, 4]. Thus, after remote exposure, IPA may no longer be detectable while acetone remains measurable.

The route of exposure in this case was unusual. Cutaneous absorption of isopropyl alcohol is recognized but underreported. Although intact skin poses relatively low risk, compromised or device-adjacent skin may permit greater transdermal absorption. Published reports include severe central nervous system depression in a child after repeated isopropyl alcohol sponging for fever control and acetone poisoning related to a synthetic plaster cast [5, 6]. Our case extends this literature by illustrating acetonemia temporally associated with topical IPA used for cleaning extracranial scalp connector sites in a neurodisabled patient with poor oral intake and limited physiologic reserve.

The patient’s elevated serum acetone concentration could reflect both exogenous exposure and concurrent ketosis. Serum acetone and beta-hydroxybutyrate concentrations vary by clinical context, and beta-hydroxybutyrate is the primary marker of ketoacidosis [7]. In this patient, beta-hydroxybutyrate was markedly elevated (7.2 mmol/L), while low BUN (3 mg/dL) and hypoalbuminemia (2.3 g/dL) supported poor nutritional status and a catabolic state. These findings favor starvation ketoacidosis as the primary driver of the anion-gap metabolic acidosis, with superimposed acetonemia likely related, at least in part, to topical IPA exposure (Table 2).

Table 2.

Expected qualitative patterns of acetone and beta-hydroxybutyrate across selected clinical contexts

Condition	Acetone pattern	Beta-hydroxybutyrate pattern
Isopropanol exposure/metabolism	May be elevated as IPA is metabolized to acetone; acetone may remain detectable after IPA falls	Not the primary abnormality unless concurrent ketoacidosis is present
Starvation ketoacidosis	May be elevated from endogenous ketone metabolism	Elevated; typically lower than severe DKA but variable by severity and duration
Diabetic ketoacidosis	Often elevated as part of ketone metabolism	Markedly elevated and a major contributor to anion-gap acidosis

One diagnostic inconsistency warrants emphasis: the markedly elevated osmolar gap (169 mOsm/kg) is not quantitatively explained by the measured acetone concentration alone. A serum acetone concentration of 29.8 mg/dL corresponds to approximately 5 mOsm/kg. Accordingly, while acetonemia supported a toxic alcohol-related diagnostic pathway, the magnitude of the osmolar gap suggests additional unmeasured osmoles, analytic variation, or other contributors not captured by the volatile panel. We therefore interpret acetonemia as a key abnormality temporally associated with topical IPA exposure, but we do not attribute the full osmolar gap to acetone alone.

Management of isopropanol or acetone toxicity is primarily supportive. Unlike methanol and ethylene glycol, isopropanol toxicity has no specific antidote, and fomepizole is generally not indicated because blocking alcohol dehydrogenase may prolong the presence of parent IPA [1, 3]. Intravenous fluid resuscitation can support renal clearance, and sodium bicarbonate may be used to buffer severe acidemia when acidosis is driven by concomitant ketoacidosis. Hemodialysis is reserved for refractory or severe cases and is not commonly required for acetone alone [1, 3]. In our case, intravenous fluids and bicarbonate therapy corrected the acidosis over 48 h, and the patient returned to baseline without renal replacement therapy.

This case underscores the importance of recognizing non-oral routes of toxin exposure, especially in neurologically impaired patients. In neurodisabled patients with device-related scalp or connector care, clinicians should specifically review topical agents used, including concentration and approximate amount; assess nutritional status as a contributor to ketoacidosis; and interpret osmolar gaps cautiously. Clinicians should maintain a high index of suspicion for transdermal toxic exposures in patients with impaired barriers and consider early measurement of serum ketones and osmolar gap in unexplained encephalopathy.

Conclusion

This case highlights a rare and clinically important presentation of systemic acetonemia temporally associated with topical isopropyl alcohol exposure in a neurologically impaired patient with device-adjacent skin care needs. The patient’s presentation with encephalopathy, high anion-gap metabolic acidosis, and an elevated osmolar gap underscores the diagnostic value of a careful exposure history, particularly when noningestional toxic exposures are possible. In this case, starvation ketosis most plausibly accounted for the anion gap, while topical isopropyl alcohol exposure with hepatic conversion to acetone likely contributed to acetonemia and supported the toxicologic diagnostic pathway. Clinicians should maintain heightened vigilance for potentially iatrogenic exposures in vulnerable populations and avoid topical isopropyl alcohol on compromised or device-adjacent skin when safer alternatives are available.

Abbreviations

ABG

Arterial blood gas

AG

Anion gap

BHB

Beta-hydroxybutyrate

BMP

Basic metabolic panel

BUN

Blood urea nitrogen

CT

Computed tomography

DKA

Diabetic ketoacidosis

ED

Emergency department

ICU

Intensive care unit

IPA

Isopropyl alcohol

VTE

Venous thromboembolism

Author contributions

N.A. and A.S. supervised and oversaw the entire study. R.A. and Q.Q. conducted the literature search, while M.Q., M.E., and M.A.B. interpreted the findings. Q.Q. and R.A. drafted the original manuscript. R.A., Q.Q., and A.S. reviewed and edited the manuscript. A.A. provided critical revisions, and A.E. contributed to writing and refining the final version. All authors read and approved the final manuscript.

Funding

None.

Data availability

Data sharing does not apply to this case report, as no datasets were generated or analyzed during the current study.

Declarations

Ethical approval

Ethics committee approval was not required for this case report under institutional policies governing single-patient case studies involving no identifiable data beyond clinical care.

Consent

Written informed consent was obtained from the patient for publication of this case report.

Provenance and peer review

Not commissioned; externally peer-reviewed

Competing interests

The authors declare no competing interests.