Blood pressure safety of subanesthetic ketamine for depression: A report on 684 infusions

Ketamine, a dissociative anaesthetic historically used in surgical settings, has attracted attention in psychiatry for rapid antidepressant effects in treatment-resistant depression and acute suicidality. Earlier literature from anaesthesia documents ketamine's sympathomimetic effects at anaesthetic doses — raising heart rate, arterial pressure and cardiac output via catecholamine release and inhibition of catecholamine reuptake — but the cardiovascular profile of subanesthetic, slow infusions commonly used in psychiatry is less well characterised. Existing pooled data from smaller trials reported larger mean blood pressure increases and occasional need for antihypertensive drugs, prompting debate about appropriate monitoring and patient selection for off-label use in mental health settings. This study presents a retrospective case series aiming to quantify blood pressure changes during routine clinical administration of subanesthetic ketamine for depression. Riva-Posse and colleagues analysed serial vital-sign data from sequential patients treated at a single centre to describe the magnitude, timing and clinical relevance of systolic and diastolic blood pressure changes during and after 0.5 mg/kg intravenous infusions over 40 minutes, and to explore patient-level predictors and repeat-infusion effects.

The investigators conducted a retrospective chart review of 66 consecutive, care-seeking adults treated in the Treatment Resistant Depression Clinic at Emory University between June 2014 and June 2016. All treatments were delivered as part of clinical care in an ECT/neuromodulation suite staffed by ACLS-certified nurses and physicians trained in moderate sedation. The clinical protocol and the retrospective analysis were authorised by institutional committees and the institutional review board. Inclusion required a current major depressive episode (unipolar or bipolar), prior failure of multiple medication trials, or consideration/receipt of electroconvulsive therapy. Exclusion criteria included absence of a primary mood disorder, active psychotic symptoms, concurrent substance use disorder, or medical instability such as uncontrolled hypertension, oxygen dependence, delirium or other conditions judged unsafe. Patients with essential hypertension were eligible if their blood pressure was controlled on antihypertensives, and each infusion required pre-infusion blood pressure below SBP 160 mmHg and DBP 100 mmHg; patients exceeding these thresholds on a given day were not infused. All patients received racemic ketamine dosed at 0.5 mg/kg in 100 ml saline infused over 40 minutes (rate 150 ml/hour). Patients fasted for at least 8 hours prior and were monitored continuously with cardiac telemetry and intermittent blood pressure readings: at baseline, at the start of infusion, every 10 minutes during the 40-minute infusion, and every 15 minutes thereafter for 30 minutes (up to 70 minutes). Infusions would be stopped for symptomatic hypertension or if SBP rose more than 45 mmHg or DBP more than 30 mmHg. Clinical assessments included the self-report QIDS-SR16 prior to each infusion and clinician-rated CGI-S and CGI-I; patients were monitored for cognitive, psychiatric and urinary side effects. The primary outcome was the change in systolic and diastolic blood pressure from pre-infusion to the highest post-infusion measurement. Initial exploratory statistics were performed with Minitab; full analyses used R v3.4.1. Between-group comparisons (for example, hypertensive versus non-hypertensive, age groups, lorazepam use) used Student's t-tests, and within-subject comparisons (for example, first versus sixth infusion) used paired t-tests. The authors also examined correlations between ketamine dose and blood pressure changes.

Sixty-six patients received 684 ketamine infusions. The cohort had a mean age of 56.7 years (SD 12.86) and was 60.6% female. Baseline depression severity (QIDS-SR16) averaged 17.38 (severe); patients had tried a mean of 4.31 antidepressant medications in the current episode and 47% had prior ECT. Most (87.9%) had major depressive disorder; eight had bipolar depression. Twenty-four patients (36.1%) had essential hypertension treated with antihypertensive medication. All infusions were completed without interruption for unstable vital signs or psychotomimetic effects, and no patient required pharmacological treatment for elevated blood pressure. Reported baseline vital-sign means were SBP 105.27 mmHg (SD 34.47), DBP 71.81 mmHg (SD 11.29) and MAP 82.96 mmHg (SD 19.02). Peak mean elevations occurred at 30 minutes (SBP 108.55 mmHg, DBP 74.98 mmHg, MAP 86.17 mmHg), corresponding to mean changes from baseline of +3.28 mmHg systolic, +3.17 mmHg diastolic and +3.21 mmHg MAP. Vital signs trended back toward baseline between the end of infusion (40 min) and the end of monitoring (70 min); the net difference between start and 70 min was small (SBP +0.6 mmHg, MAP +1.56 mmHg, DBP +2.03 mmHg reported in different places in the text). Analyses of the 41 patients who received at least six infusions showed no statistically significant change in blood pressure measures between the first and sixth infusion at any measured timepoint (start, 30 min, 40 min, 70 min), indicating no clear development of tolerance to the pressor effect over repeated exposures. Across all infusions, 63 instances (9.2% of infusions) met a predefined criterion for clinically significant elevation (systolic > 30 mmHg or diastolic > 15 mmHg); 50 of these events (79.5%) were primarily diastolic elevations, and 78% of the significantly elevated readings occurred at the 30- or 40-minute marks. Thirty-three patients (50% of the sample) experienced at least one infusion with such an elevation, but only a single subject had five or more such events; the investigators note these elevations were sporadic rather than recurrent within individuals. Dose analyses showed all patients received 0.5 mg/kg (mean 41.25 mg, SD 10.34; range 20–61 mg). Ketamine dose correlated positively with change in DBP (R = 0.106, P = 0.010) but not with change in SBP. Additional subgroup findings reported in the Discussion include that hypertensive patients had higher peak pressures during infusions and older patients (age ≥ 60) exhibited larger increases in systolic pressure and smaller diastolic increases at 30 minutes, but no statistical details beyond those summaries are provided in the extracted text. The authors also report that patients who received lorazepam immediately prior to infusion had lower absolute blood pressure readings across the infusions but did not show attenuated rises from start to peak.

Riva-Posse and colleagues interpret their findings as evidence that subanesthetic ketamine administered at 0.5 mg/kg over 40 minutes produces small, transient increases in systolic and diastolic blood pressure that are statistically detectable but clinically modest. With mean rises of roughly 3.2 mmHg in SBP, DBP and MAP at peak, and return toward baseline by 70 minutes, the authors argue these haemodynamic changes are unlikely to be clinically meaningful in most patients. The inclusion of patients with treated essential hypertension and the finding that hypertensive patients had higher peaks that resolved by the end of monitoring led the investigators to suggest that well controlled hypertensive patients need not be automatically excluded from receiving ketamine infusions. The authors note several subgroup observations: patients aged 60 and over showed larger systolic but smaller diastolic increases at 30 minutes, a pattern attributed to age-related decreased vascular compliance; lorazepam given prior to infusion was associated with lower absolute blood pressure readings but did not blunt the increase from baseline to peak, which could reflect anxiolytic lowering of baseline pressure without blocking ketamine's pressor reactivity. They also report no evidence of tolerance to the blood pressure effects across up to six infusions and emphasise that clinically significant elevations were uncommon (under 10% of infusions) and usually sporadic. Key limitations acknowledged include the retrospective, single-site design and use of a convenience clinical sample rather than participants enrolled under trial-standardised procedures. The extracted data lacked standardised measures of dissociation, limited racial and ethnic diversity, and did not permit correlation of blood pressure changes with antidepressant clinical outcomes. The authors caution against generalising these results to other doses, infusion rates, routes of administration, or to populations such as recreational users or patients with unmanaged comorbidities. They call for further research on long-term cardiovascular, cognitive, genitourinary and hepatic safety as well as abuse liability. Overall, the investigators conclude that in this clinical series ketamine at 0.5 mg/kg infused over 40 minutes was well tolerated from a blood pressure standpoint in the acute care setting, providing some reassurance about its cardiovascular safety in similar clinical populations.