class: RWH_bg_title # .black[HypoK in the Receiving Unit] ### Medicine24 .RWH_footnote_title[ .RWH_footer_bold[ Rob Hunter | @renalrob ] ] .RWH_footnote_right_title[ .RWH_footer_bold[ 27th Oct 2022 ] ] ??? Medicine 24 27th Oct 2022. Slides created with [xaringan](https://github.com/yihui/xaringan). Image credits: - [ECG](https://casereports.bmj.com/content/2009/bcr.07.2008.0577.long) - [Chimp](https://www.pexels.com/photo/view-ape-thinking-primate-33535/) - [Lizard](https://www.pexels.com/photo/brown-and-green-lizard-584165/) --- # These slides These slides complement the presentation given at Medicine 24. Press 'p' to see speaker notes, with additional information. ??? ---  ??? # The "med reg" approach Hypokalaemia is common (1/6th of medical admissions) and usually doesn't require a lot of thought. 98% of cases are caused by GI losses or diuretics. EtOH often implicated too. Therefore, the medical SpR ends up using their reptilian brain. A sensible adaptation in many ways because this will work 98% of the time. However, the key is to know *when* to pause and ["think slow"](https://en.wikipedia.org/wiki/Thinking,_Fast_and_Slow) - i.e. when to use our mammalian brain. ## Some background In a [Danish cohort study](http://dx.doi.org/10.1016/j.amjmed.2014.07.022) of 12,000 acute medical patients: - hypoK in 1/6 of admissions (RFs elderly, women, cancer, EtOH, diuretics) - K < 2.9 mM associated with doubling of early mortality The mortality risk may be due to confounding (e.g. sick patient, adrenaline surges) - but is in part probably due to suppressed NaKATPase activity, resulting in [intracellular Na and then Ca overload](https://www.ahajournals.org/doi/10.1161/circulationaha.115.016217) in ventricular cardiomyocytes. ---  <!--[:scale_c 80%]--> ??? # The medical student approach If you send a medical student to clerk in a patient with hypokalaemia then they will - quite rightly - use this sort of algorithm. Two problems: 1) this is too complicated and cognitively cumbersome: we don't have time to quiz 1/6th of all admissions about licorice ingestion and glue-sniffing 2) relies on 24 hr urines (never present) or spot UK/Cr (rarely present and not within first 24 hrs); by the time the medical student has diagnosed a VIPoma, the patient has already been discharged on Sando-K <br> Algorithm re-construced from [Unwin, Luft & Shirley](https://www.nature.com/articles/nrneph.2010.175), [Gennari](https://www.nejm.org/doi/full/10.1056/nejm199808133390707), [Weiner & Wingo](https://pubmed.ncbi.nlm.nih.gov/9219169/) & [UpToDate](https://www-uptodate-com.ezproxy.is.ed.ac.uk/contents/causes-of-hypokalemia-in-adults). NB also mechanisms of [EtOH](https://www.nejm.org/doi/full/10.1056/NEJM199312233292605) and [cola-induced](https://pubmed.ncbi.nlm.nih.gov/23186978/) hypoK. <br> <br> ## Some notes on miscellaneous causes: - low-calorie diet = hypoK - probably due to ketosis (mechanism not clear) - EtOH induces a reversible Fanconi syndrome (+/- ketosis) <br> ### Hypokalaemic period paralysis: - may be inherited (FPP) or acquired in thyrotoxicosis (TPP) - FPP due to AD mutations in SkM Ca or Na channels - TPP common if East Asian family origin; commonly find low PO4 (Lin, QJM 2001) - ...and in both cases precipitated by exercise, CHO load, stress <br> ### Aminoglycosides can cause: - non-oliguric AKI (10 – 20 %) – RFs female, hypovolaemia, CKD, duration, other nephrotoxins, hypoK, hypoMg - proximal syndrome (Fanconi) = acidosis - distal syndrome (Bartter) = alkalosis Post-gent Bartter = pretty rare (2 – 4 %); more common in females; normal Cr; FEMg (remember alb) and FECa (CCCR) helpful; persists for up to 6 weeks. The mechanism is said (e.g. [Kamel & Halperin, 2021](https://doi.org/10.1016/j.ekir.2021.02.003) to be that cationic aminoglycosides activate the CaSR in the LOH - and can get similar effect with cisplatin, amikacin and cationic paraproteins. --- # Overview - two cases: - hypokalaemia in a man with a new cancer diagnosis - hypokalaemia and paralysis in a young woman <br> - clinical pearls for the reptilian physician: - high-yield investigations - how to make a diagnosis in the first 24 hrs - tips and pitfalls with K⁺ replacement .RWH_footnote_right[.RWH_footer_style[slides at: www.kidneyfish.net/talks/]] ??? --- class: center, middle, inverse # .white[Case 1] ### HypoK in a 62M with a new diagnosis of cancer [ case details redacted for online slides ] ??? On the face of it this looks like vomiting. But look - urinary K excretion is high - so what is going on? ## Learning points (diagnosis): - urinary K losses after vomiting - value of assessing chloride (blood or urine) - extreme alkalosis is always vomiting - very extreme alkalosis = pyloric stenosis or Zollinger-Ellison syndrome (Galla) <br> What is the mechanism that drives kaliuresis after vomiting? See [Nephromaths eBook](https://kidneyfish-nephromaths.netlify.app/general-approach-to-a-patient-with-electrolyte-disturbance.html#vomit). <br> ## Learning points (replacement): - use IV if severe or route unavailable - usually KCl as chloride deficiency prevalent and also helps to keep it in extracellular space --- class: center, middle, inverse # .white[Case 2] ### Hypokalaemia with paralysis in a 28F [ case details redacted for online slides ] ??? Profound paralysis - persisted for 2 weeks on ITU. ## Learning points (diagnosis): - value of old labs - very hard to drop K by dietary restriction alone - acidosis = RTA or diarrhoea - not all paralysis is HPP - acid-base status normal in HPP <br> ## Learning points (replacement): - dietary fortification usually ineffective - if acidotic then need base (e.g. K citrate) ---  ??? First take a history and exam, paying particular attention to GI losses (quantify), volume status and ABP. Differentiate by: - acid-base status (including chloride) - other electrolytes - urinary electrolytes - but results won't be back immediately If FPP, acid-base status always normal and TTKG always < 3 and UK/Cr < 2.5 ([Lin et al., QJM 2001](https://academic.oup.com/qjmed/article/94/3/133/1524035) and [AIM 2004](https://pubmed.ncbi.nlm.nih.gov/15277290/)). In "secondary" hypoK paralysis, an acid-base disorder is invariably present. "Pan-hypo-electrolytes": - tubular salt-wasting (including DKA) - GI losses - intracellular shifts (re-feeding) - hypoMg and hypoK exacerbates renal K wasting ---  ??? Reasonable thresholds: - **renal losses** likely if... UK:Cr > 1.5 (or 2) mmol / mmol | UK > 20 mM | > 20 mmoles / day | TTKG >> 3 | FEK > 10 – 15% - **appropriate renal response** if... UK:Cr < 1.5 (or 2) mmol / mmol | UK < 15 mM | < 15 mmoles / day | TTKG < 2 | FEK < 2 – 3% - **low urinary chloride** = UCl < 10 mM <br> This is easiest to remember as "all the 2s" - UK:Cr 2; TTKG 2; spot UK 20; 24 hr UKER 20. --- # Intravenous potassium - for severe hypokalaemia or if oral route not available - prefer KCl in normal saline <br> - usually 10 – 20 mmol per hr - *in extremis* up to 40 mmol per hr (central line on ITU) - don’t exceed 40 mM by peripheral vein <br> - total deficit 100 mmoles per 0.3 mM drop in [K⁺]… - …but careful if transcellular shifts - expect 0.1 mM increment per 10 mmol ??? - IV replacement if K < 2.5 mM or arrhythmia - give in saline (NOT dextrose - to avoid shift into cells) - correct Mg deficiency - risk of rebound with distributive hypoK - with chronic K losses, estimate total K deficit as 100 mmoles per 0.3 mM (so K < 3 = 200 mmol; K < 2 = 1000 mmol) - [Asmar et al., AJKD 2012](https://pubmed.ncbi.nlm.nih.gov/22901631/). --- # Oral potassium - dietary advice usually insufficient - choice of anion (chloride or bicarbonate) is important <br> - all formulations have GI side-effects - remember amiloride (or spironolactone) <br> - maintenance requirement = 60 mmol - K > 3.0 give 75 mmol in divided doses (e.g. Sando-K TT tds) - K < 3.0 give 100 mml in divided doses (e.g. Sando-K TTT tds) - prescribe no more than three days to start off ??? - prefer oral for most - usually give with chloride (due to prevalence of chloride deficiency and will correct faster; chloride stays in extracellular space) - give with chloride if alkalosis - give with HCO3 or precursors if acidosis - in food, mainly with phosphate or citrate (so a lot goes into cells) - consider amiloride (or MRA if mineralocorticoid XS) - all associated with GI symptoms (direct irritation by K and also effervescent preparations can cause bloating) - risk of intestinal perforation with slow-release tablets - sandoK = KHCO3 + KCl - Kay-Cee-L = KCl - K citrate = liquid or effervescent tablets - K acid tartrate with KHCO3 <br> <br> ## Source Summary of guidelines from units around UK, plus [UpToDate](https://www-uptodate-com.ezproxy.is.ed.ac.uk/contents/clinical-manifestations-and-treatment-of-hypokalemia-in-adults), [KDIGO](https://doi.org/10.1016/j.kint.2019.09.018), [Kruse & Carlson](https://jamanetwork.com/journals/jamainternalmedicine/fullarticle/612991) & [Cohn et al.](https://pubmed.ncbi.nlm.nih.gov/10979053/): - mild (K > 3.0) = sandoK TT tds = 72 mmol or Kay-Cee-L syrup 25 ml tds = 75 mmol (or Slow-K but risk of intestinal perforation with slow-release preps; also tablets are very large) - severe (K < 3.0) = sandoK TT qds = 96 mmol of Kay-Cee-L 25 ml qds = 100 mmol - IV - give 20 -- 40 mmol in 1L over at least 2 hrs - and some say up to 4 or 6 hrs (or in 100 ml centrally if in critical care on monitor); some use 20 or 40 mmol in 500 ml; give 40 mmol in 1L bd or tds - don't exceed 20 mmol/hr (and usually not 10 mmol/hr) - NB think this probably comes from Kruse et al (AIM 1990): observational study of 1350 infusions in 190 patients; 20 mmol infusion induced mean 0.25 mM increment; rates of 20 mmol/hr were safe - cardiac monitor if exceeding 20 mmol/hr - don't exceed 40 mM by peripheral cannula (venotoxic) - daily maintenance = 40 -- 80 mmol; maximum is 2 -- 3 mg/kg - can give s.c. for maintenance - high risk in digoxin toxicity - caution with K replaceament if AKI / CKD (and never replace if immediately post-HD) - consider K-sparing diuretic as alternative - prefer pre-mixed bags - prescribe 3 days at a time max - sandoK may not be appropriate if significant metabolic alkalosis - dietary replacement alone unlikely to work (as KPO4 won't replace chloride losses from diuretics / GI losses) - anecdotally: Kay-Cee-L better tolerated than sandoK or citrate (bloating issue with sandoK); take citrate with meals helps --- # HypoK in the first 24 hrs - golden diagnostic window (send spot urine sample) - high-yield tests: - blood pressure - TCO₂ (HCO₃) and chloride - historic data - vomiting causes urinary K⁺ wasting - choose anion wisely in replacement .RWH_footnote_right[.RWH_footer_style[slides at: www.kidneyfish.net/talks/]]