Beyond static thresholds: the “citrate challenge” as a physiology-guided bedside framework

Dear Editors,

We thank Müller and David for their thoughtful comment on our editorial on the “citrate challenge” [1, 2]. We fully agree with their central point that citrate accumulation (CA) still lacks a universally accepted and clinically meaningful definition, and that harmonized criteria remain needed. However, we believe their interpretation partly differs from the intent of our proposal. The citrate challenge was not conceived as a score designed to predict CA from a single baseline variable. Rather, it was proposed as a pragmatic bedside framework for managing uncertainty in patients with reduced metabolic reserve. Much as in hemodynamic practice, a “challenge” is not meant to compress physiology into a single number, but to support a monitored and adaptive approach when tolerance is uncertain [3].

First, step 1 of our approach was not intended to promote vasopressor requirement as a standalone screening test for CA. It was simply meant to identify clinical contexts in which citrate metabolism may be limited from the outset, such as severe circulatory shock, advanced liver dysfunction, or intoxications affecting the tricarboxylic acid cycle [4]. In that respect, the poor discriminative performance of the vasoactive inotropic score (VIS) reported by Müller et al. should not be viewed as contradictory [5]. Indeed, even in their cohort, baseline VIS remained numerically higher in patients who developed CA than in those who did not. The purpose of this first step was therefore not to replace monitoring, but to anticipate which patients warrant heightened vigilance from the very start of RCA.

Second, the core of the citrate challenge is not using a baseline threshold, but a dynamic bedside assessment. Clinically relevant CA is a dynamic syndrome. The meaningful question is not whether a single value crosses a cut-off at one time point, but whether a coherent pattern emerges over time: rising tCa/iCa ratio, persistent or worsening ionized hypocalcemia despite calcium substitution, metabolic acidosis with increasing anion gap, and failure of the overall metabolic profile to improve despite resuscitative measures [6–8]. This is precisely why we advocated repeated evaluation every 3–4 h and emphasized that the association and kinetics of markers matter more than any isolated measurement [1, 6].

Third, our proposal to consider lactate > 10 mmol/L or pH < 7.15 was not intended to define the “optimal” threshold for predicting CA in all patients. It was intended to identify a subgroup at risk of very early CA, that is, patients in whom the time available to react may be particularly short. This distinction is important. Müller and David appropriately report that a lactate cut-off of 2.6 mmol/L, derived from the Youden index, yielded only moderate sensitivity and specificity in their cohort [5]. In our view, this highlights a broader mathematical limitation of relying on baseline lactate alone. When early CA is relatively rare, a high negative predictive value is mostly expected from the low prevalence of the event and should not be overinterpreted as evidence of strong bedside discrimination. From a bedside perspective, the question can be framed in Bayesian terms: does the observed baseline value meaningfully shift the estimated probability of early clinically relevant CA enough to modify management? In that regard, likelihood ratios are particularly informative, because they quantify the change from pre-test to post-test probability. A low lactate threshold chosen to maximize sensitivity will inevitably reduce specificity and label many patients as potentially “at risk,” while often producing only modest positive likelihood ratios. Used as a gateway test, such a threshold could therefore lead to withholding RCA in many situations where clinically relevant early CA would never have occurred. Our intent was therefore not to identify a perfect marker, but to stratify the urgency of risk and adapt the intensity of monitoring and treatment accordingly. This was also the rationale for proposing, in experienced centers, a protocolized low-citrate strategy under close monitoring, while acknowledging that such an approach may involve trade-offs, particularly with respect to acid–base control [9].

Fourth, the citrate challenge was also designed to promote a combined evaluation of biochemical markers, rather than reliance on the tCa/iCa ratio alone. In our view, this ratio should always be interpreted in the context of the concurrent evolution of ionized calcium, acid–base status, anion gap, calcium substitution requirements, and the patient’s overall response to resuscitation. We likewise believe that albumin-corrected calcium ratios should not be used as standalone evidence of CA [10]. Such corrections may amplify a signal already influenced by several non-citrate determinants, thereby moving interpretation further away from a physiology-based assessment of the whole phenomenon.

Finally, we agree with Müller and David that more elaborate and validated models may improve risk estimation in the future. However, we would caution against replacing one oversimplification with another. The “challenge” is not to find a single perfect marker, but to develop tools that remain faithful to physiology and useful at the bedside. In that regard, future work could move beyond fixed cut-offs and explore staged definitions of CA integrating both the trajectory of the tCa/iCa ratio and the concurrent evolution of hypocalcemia, acid–base abnormalities, and calcium requirements. Such an approach may better capture the continuum from biochemical warning signs to clinically relevant CA.

In summary, we believe the message of the “citrate challenge” remains unchanged: monitor regularly, interpret markers dynamically, anticipate early deterioration, and adapt RCA to the patient’s trajectory rather than to a single number. If our editorial has contributed to shifting the discussion from static thresholds toward dynamic physiological assessment, then the exchange initiated by Müller and David will have been a valuable step forward.

Author contributions

All authors were involved in drafting the manuscript or revising it critically for important intellectual content. All authors read and approved the final version of the manuscript.

Funding

None.

Data availability

No datasets were generated or analysed during the current study.

Declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.