Is the recommendation not to use rifampin plus pyrazinamide for latent tuberculosis treatment always imperative?

saline solution bolus via a central venous catheter, a thermistor-tipped arterial catheter (usually femoral) is connected to a specific monitor (PiCCOplus; Pulsion Medical Systems; Munich, Germany), which records the downstream temperature changes, ie, a transpulmonary thermodilution curve. The mathematical analysis of the curve allows the computation of cardiac output, heart blood volume (global end-diastolic volume), and an estimation of extravascular lung water. 1– 4 Recently, the mere observation of the transpulmonary ther-modilution curve has been shown to be useful to detect and monitor right-to-left intracardiac shunting. 5 Indeed, in case of right-to-left intracardiac shunt, one part of the cold indicator passes through the atrial septum and rapidly reaches the arterial thermistor. As a result, the thermodilution curve appears prematurely and becomes biphasic with two humps (" camel " curve). 5 I would like to report such a camel thermodilution curve that was not related to a right-to-left intracardiac shunt. This curve was recorded in a patient with septic shock instrumented with a right femoral thermistor-tipped arterial catheter and a right femoral venous catheter. Both catheters were of the same length (20 cm). When the cold saline solution was injected through the femoral venous line, the transpulmonary thermodilution curve was highly suggestive of right-to-left intracardiac shunting (Fig 1). However , the patient was not hypoxemic and contrast echocardiography as well as color Doppler imaging did not reveal any intracardiac shunt. Because both arterial and venous femoral catheters were on the same side and of the same length, it was hypothesized that the high concentration of cold indicator at the site of injection (the tip of the venous catheter) may induce significant temperature changes in the femoral artery, ie, at the level of the arterial catheter tip equipped with a thermistor (cross-talk phenomenon). To confirm this hypothesis, central venous injections were also performed through an internal jugular line. In this case, the shape of the transpulmonary thermodilution curve was normal (Fig 1). Looking at transpulmonary thermodilution curves can be very useful to diagnose intracardiac shunts. 5 However, a cross-talk phenomenon may result in a double-hump thermodilution curve wrongly suggestive of right-to-left intracardiac shunting. Therefore , the use of venous and thermistor-tipped arterial catheters on the same side and of the same length should be avoided in patients monitored with transpulmonary thermodilution.