The use of Tablo continuous veno-venous haemodialysis to rapidly remove lithium in a patient with severe lithium toxicity

  1. Brian Nohomovich ,
  2. Emmanuel Tito ,
  3. Michael Terrio and
  4. Matthew Belardo
  1. Internal Medicine, Western Michigan University Homer Stryker MD School of Medicine, Kalamazoo, Michigan, USA
  1. Correspondence to Dr Brian Nohomovich; bnohomovich@gmail.com

Publication history

Accepted:04 May 2023
First published:15 May 2023
Online issue publication:15 May 2023

Case reports

Case reports are not necessarily evidence-based in the same way that the other content on BMJ Best Practice is. They should not be relied on to guide clinical practice. Please check the date of publication.

Abstract

We report a case of severe, life-threatening lithium toxicity in a patient with bipolar I disorder who presented with altered mental status and acute renal failure. At admission, serum lithium level was well above toxic levels (>2 mEq/L). The signs and symptoms of lithium toxicity significantly improved after treatment with continuous veno-venous haemodialysis (CVVHD). The patient was discharged with no neurological or renal sequelae. Herein is the first case report using the Tablo CVVHD system to treat severe lithium toxicity.

Background

Lithium is a first-line medication for treating bipolar disorder, recommended by the National Institute for Health and Care Excellence guidelines.1 Lithium is used to manage other medical conditions, such as refractory depression, cluster headaches and impulsive behaviour.2 3 Lithium intoxication occurs at various degrees of severity due to lithium’s narrow therapeutic window, which can lead to severe multiorgan dysfunctions, ultimately leading to the patient’s death. The clinical manifestations of lithium intoxication can be categorised as acute, acute-on-chronic or chronic.4 Early diagnosis and treatment of toxic lithium poisoning remain essential in preventing morbidity and mortality. There is still debate about the preferred extracorporeal treatment modality for severe lithium intoxication. Because lithium is readily diffusible, intermittent haemodialysis (IHD) is preferred.5 However, this could lead to post-dialysis rebound concentration and aggravation of haemodynamic instability. Continuous renal replacement therapy (CRRT) may be valuable due to its continuous solute removal.5 We present and discuss the first case of severe lithium toxicity successfully treated with continuous veno-venous haemodialysis (CVVHD) using the Tablo haemodialysis system.6

Case presentation

Our patient, with a medical history of bipolar disorder, hypertension, non-alcoholic fatty liver disease, anxiety, depression and early-onset dementia, presented with worsening mental status. She has been taking lithium for over 18 months and adheres to medication routinely prescribed by her primary care provider. She had laboratory work every 3 months in the outpatient for lithium monitoring, assessing renal function, thyroid function, serum lithium levels and urinalysis. All laboratory results before admission were within normal limits. The patient reported acute confusion 5 days before admission and felt that she was not acting herself. She denied any precipitating illness. She developed intractable muscle movement and upper extremity tremors 4 days before admission. Her confusion continued to progress, and she was having difficulty with ambulation. Three days before admission, she visited a nearby hospital emergency department (ED), and her lithium level was found to be 3.3 mEq/L (0.6–1.2 mEq/L) and creatinine at 1.4 mg/dL (0.6–1.1 mg/dL; baseline 1.0 mg/dL). She was advised to cut her lithium dose in half from 450 mg four times a day to 450 mg two times per day. She reported that she followed those guidelines. Over the next 2 days, after adjusting her dose regimen, she reported progressively worsened confusion, fatigue, intense tremors, poor appetite, visual hallucinations, butterflies, decreased urine output and muscle aches. On additional review of systems, she reported headaches, decreased concentration, sleep disturbances and nausea. The patient presented to our ED haemodynamically stable with normal blood pressure and pulse range. She had an SPO2 of 100% on ambient air with a respiration rate of 16. She did not appear in acute distress and was comfortable in the hospital bed. ECG showed normal sinus rhythm, no axis deviation, and no ST elevations or depressions. Her examination findings include myoclonus in both upper and lower extremities and generalised muscle tenderness to palpation. The rest of the neurological examination was normal, with no focal deficits. Cardiovascular, respiratory and abdominal examinations were unremarkable.

Differential diagnosis

The patient’s initial clinical presentation showed that lithium toxicity was on top of the differential. Other potential explanations for the patient’s symptoms at the time of presentation included a posterior fossa cerebrovascular event, substance abuse and Wernicke syndrome, all being less likely given the clinical history, lack of focal neurological deficits on the examination, laboratory and imaging results. Our patient presented with decreased renal function, suspected of chronic lithium exposure. Hypercalcaemia was noted on admission laboratory work and could be due to lithium toxicity. Elevated serum lithium level confirmed and established the definitive diagnosis for this patient.

Investigations

At admission, laboratory studies showed mild neutrophilic leucocytosis with white cell count of 15.0×109/L (4–11×109/L), decreased serum sodium 134 mmol/L (135–145 mmol/L), elevated serum creatinine 2.27 mg/dL (0.6–1.1 mg/dL), glomerular filtration rate 24 mL/min, (>60 mL/min), elevated calcium 12.3 mg/dL (8.6–10.3 mg/dL), adjusted calcium 11.5 mg/dL, serum albumin 5.0 (3.5–5.0 g/dL), serum phosphorus 1.9 (2.7–4.5 mg/dL), serum magnesium 2.0 (1.7–2.8 g/dL), serum osmolality 311 mOsm/kg (275–295 mOsm/kg), elevated thyroid-stimulating hormone (TSH) of 10.10 µ/mL (0.27–4.2 µ/mL) and elevated lithium level 5.1 mEq/L (0.6–1.2 mEq/L). Urinalysis showed mild ketonuria with a negative urine culture. Additional random urine studies showed urine creatinine 16.7 mmol/L (7–14 mmol/L), urine osmolality 477 mOsm/kg (300–1100 mOsm/kg), urine nitrogen 612 mg/dL (28–714 mg/dL) and urine sodium 78 mmol/L (20–40 mmol/L). Based on these laboratory results, nephrogenic diabetes insipidus (NDI) was ruled out. Chest radiograph showed no acute process for cardiopulmonary disease.

Treatment

In our ED, she received 2 L of normal saline for hypercalcaemia and acute kidney injury and started on maintenance normal saline. The patient was appropriate for intermediate haemodialysis, given that she was haemodynamically stable. However, intermediate haemodialysis was unavailable due to a shortage of medical staff trained to operate the haemodialysis machines. After consulting poison control, they recommended CRRT as an alternative. She was immediately transferred to the medical intensive care unit (ICU) for emergent Tablo CVVHD for lithium toxicity and subacute renal failure. The Tablo haemodialysis system was run with the following settings (treatment time: 22 hours, blood flow rate 400 mL/min, dialysate flow rate 90 mL/min, dialysate potassium 4 mEq/L, dialysate calcium 2.25 mEq/L, total buffer setting 35 mEq/L, sodium setting 140 mEq/L, dialyser type Optiflux F160NR, fluid removed −1.0 L, litres processed 200.00 L). We used renal function panel 401 for replacement fluid (potassium 4 mEq/L and calcium 3 mEq/L). The total run time was 22 hours. The patient improved dramatically both in mentation and clinical examination findings. She was alert and oriented three times with decreased confusion within 8 hours of being on CVVHD. She did not report any other hallucinations. On physical examination, there was resolution in her tremors and clonus. She was transferred to the medical floor the following day after treatment.

Outcome and follow-up

We monitored lithium levels and observed a rapid correction in the lithium levels in this patient (figure 1). We had an initial lithium concentration of 5.1 mEq/L just before initiation, which reduced to 2.2 mEq/L within 4 hours and then within normal levels by 14 hours at 0.6 mEq/L. Serum lithium level at 28 hours was 0.5 mEq/L (CVVHD was discontinued 6 hours prior). Subsequent daily checks over the next 2 days did not show a rebound in lithium levels. We did have to replete phosphate while on CVVHD. We rechecked her TSH, which decreased to 4.79 µ/mL, free thyroxine 1.5 µ/mL and creatine kinase 211 U/L (reference range <172 U/L). Following treatment with CVVHD, all of her symptoms had resolved by discharge. She underwent physical and occupational therapy while in the hospital, gradually returning to her baseline activity. Because cessation of lithium carbonate treatment could raise the patient’s risk of recurrence of bipolar disorder, we advised the patient to follow up with a psychiatrist outpatient. When the patient was discharged, her renal function was at baseline.

Figure 1

Timeline of the lithium level. The blood concentration of lithium declined over time. CVVHD, continuous veno-venous haemodialysis.

Discussion

The toxic effects of lithium are well documented and commonly include renal and thyroid toxicity. Lithium is a small molecule known to be nephrotoxic, with adverse effects occurring within a few days of use. The most common side effect noted is polyuria, with ~20% developing NDI.7 Lithium-induced NDI is a complex process that presents acutely with polyuria, hypovolaemia, electrolyte disturbances and chronically with decreased glomerular filtration rate.7 In NDI, aquaporin 2 channels are downregulated, causing reduced water reabsorption, increased urine output and salt-wasting.8 Aquaporin 2 channels are water-selective transport channels on the apical plasma membrane of principal cells in the kidney collecting duct.9 Vasopressin upregulates these channels, which is imperative for water balance.9–11 The mechanism of action of lithium-induced NDI is multifactorial and still not completely understood. However, epithelial sodium channels (ENaC) play a large part by facilitating lithium entry into the principal cells. In one study, αENaC were knocked out in mice, leading to only mildly reduced expression of aquaporin 2 channels when exposed to lithium, and the mice did not develop NDI.12

While our patient did not develop NDI, she did develop thyroid dysfunction. Lithium has many effects on the thyroid: it impairs iodine uptake, impairs tyrosine iodination, alters thyroglobulin structure and impairs the release of thyroxine from the thyroid.13 These effects lead to hypothyroidism. Lithium can affect the parathyroid glands as well.

Lithium can affect parathyroid hormone (PTH) secretion and cause PTH-dependent hypercalcaemia14 and subsequent hyperplasia of the parathyroid gland.15 Lithium likely interacts with the calcium-sensing receptor16 through an unknown mechanism leading to a signalling cascade resulting in a shift in the set point of calcium concentrations relative to PTH. Parathyroid chief cells then secrete PTH in response to the new set point.14 The prevalence of lithium-associated hypercalcaemia was estimated to be 26% based on a study of patients with bipolar on lithium.17 The serum hypercalcaemia in our patient resolved with fluid administration and remained within the normal range for the remainder of her hospitalisation. She was seen routinely by a provider in the outpatient before her hospitalisation, and her serum calcium levels were within normal limits. If she had persistent hypercalcaemia, a workup with PTH and ultrasound to assess the parathyroid glands is warranted, given the effects of lithium on calcium regulation.

Our case is the first case report to use Tablo CVVHD to correct acute, severe lithium toxicity rapidly. The Tablo haemodialysis system has Food and Drug Administration authorisation for use in acute and chronic care facilities and home use.6 Tablo’s maximum dialysate flow rate is 300 mL/min. Traditionally, flow rates for dialysis machines approach 800 mL/min.18 19 Recent studies have shown that slower dialysate rates are equally effective in more advanced dialysers achieving dialysis goals.18 20 Appropriate solute clearance and fluid goals are achieved in acute and chronic environments, with Tablo using a mid-dialysate flow rate close to 300 mL/min.21 Tablo has the flexibility to provide IHD or CRRT.

There are many different modalities of extracorporeal filtration, which can be separated into two main categories of renal support: continuous and intermittent. CRRTs are performed 24 hours/day, while IHD is performed 3–5 hours per session. All modalities use similar extracorporeal blood circuits; however, the difference lies between the ultrafiltration rate and whether solutes are removed by diffusion or convection.22 IHD is recommended for most toxicities as it has the highest removal rate. The downside to IHD is possible haemodynamic instability with removing large fluid volumes. We could not use IHD at the time due to staffing shortages to run IHD. Our nursing staff in the ICU was trained to operate Tablo CRRT. CRRT can and has been used to remove toxic substances. The different types of CRRT include continuous veno-venous haemofiltration (CVVH), CVVHD and continuous veno-venous haemodiafiltration. CVVH uses ultrafiltration with a high rate across a semipermeable haemofilter, which causes solute transport by convection.22 In convection, solute movement is limited due to the size of the haemofilter membrane pores. As the size of the solute molecules approaches the size of membrane pores, solute movement decreases. Due to the high ultrafiltration rate, it may be necessary to supplement with intravenous crystalloids to acquire the desired net fluid balance. CVVHD uses a dialysate that is perfused across a dialysis membrane where the primary rate of solute clearance is through diffusion down a concentration gradient. Diffusion is very effective at removing small molecular weight molecules (<500–1500 Daltons) like lithium; however, it is ineffective at removing larger molecules.22 Because there is no need for high fluid flow rates, it is easier to permit a net negative fluid balance without replacing intravenous fluids.

The case report here is helpful for centres looking to treat lithium toxicity. Case reports have shown that CVVH can fail to clear lithium, requiring IHD.23 However, our case presents that CVVHD can be an alternative when IHD is not readily available to remove lithium.5 The decision to use IHD or CVVHD should be based on the patient’s clinical status and resources in the local healthcare system. Tablo has been used recently to treat severe methanol poisoning.24 Additional investigations are needed to determine if Tablo can clear other small-molecule compounds.

Learning points

  • Tablo continuous veno-venous haemodialysis can be used to correct lithium toxicity rapidly.

  • Lithium toxicity can affect thyroid, parathyroid and renal function.

  • Suspect lithium toxicity in patients with altered mental status and acute renal failure.

  • Hypercalcaemia can also cause altered mental status and impaired renal function.

Ethics statements

Patient consent for publication

Acknowledgments

Ann Harris and the staff at the WMED library helped us with the literature review. The hospital staff helped us with this patient while she was in the ICU.

Footnotes

  • Contributors BN is the lead author and is responsible for the design of the case report, planning, drafting, and writing of the case, research, and final editing. ET was part of the team that oversaw the patient, conducted research on lithium toxicity treatment and drafted the sections regarding lithium toxicity. He also directly edited the overall final draft. MT directly researched the lithium toxicity mechanism in the discussion. He drafted that part of the manuscript and participated in the final editing process. MB directly oversaw the patient, drafted the learning points, and participated in editing the overall document and final edits. All authors participated in the final editing and agree to assume responsibility for the accuracy of this document.

  • Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.

  • Case reports provide a valuable learning resource for the scientific community and can indicate areas of interest for future research. They should not be used in isolation to guide treatment choices or public health policy.

  • Competing interests None declared.

  • Provenance and peer review Not commissioned; externally peer reviewed.

References

Use of this content is subject to our disclaimer