Abstract Background Paediatric cardiology presents unique challenges with its diverse and complex cases, limited evidence base, and the necessity for multi-expert involvement in decision-making processes. In this context, the introduction of generative pre-trained transformer (GPT) based large language models (LLMs) offers a potential avenue for the provision of complex information and clinical decision support. Purpose This study evaluates the quality of three different GPT LLMs in answering complex medical questions, including a state-of-the-art preview model that incorporates the German paediatric cardiology guidelines. Methods Seven paediatric cardiologists and paediatric cardiac surgeons generated 72 questions, including complex questions and medical cases with associated questions. The questions were categorized by difficulty and required knowledge (factual and experience-based or mostly experience-based). We prompted the questions to three LLMs: GPT 3.5, GPT 4 and a GPT 4 turbo preview. The GPT 4 turbo preview was customized by incorporating all guidelines from the German Society for Paediatric Cardiology by a retrieval function. Employing one complex instruction for all questions, we prompted the LLMs to provide precise and detailed expert-level responses. The responses from each model were evaluated by experts based on relevance, factual accuracy, severity of possible harm, completeness, superfluous content, and age-related appropriateness from 0 (very bad) to 7 (very good). Differences were calculated using the Kruskal-Wallis-test in SPSS Version 28. Results Our findings indicated a good performance of all models regarding the dimensions tested. The figures show the average ratings (Figure 1, Figure 2A) and highlight significant differences after Bonferroni correction in bold (Figure 2B). The GPT 4 turbo preview, including the retrieval of guidelines, provided significantly more relevant (average rating [AR] 5.94, meaning mostly relevant), accurate (AR 5.6, meaning between somewhat and mostly accurate) and complete (AR 5, meaning fairly complete) answers compared to GPT 3.5 and GPT 4. In terms of difficulty levels or the type of questions, there was no significant difference in rating. Relevance ratings were slightly better in factual questions (AR 5.7) than in those requiring more experience-based knowledge (AR 5,3). Although GPT4 had higher average scores compared to GPT 3.5 in all dimensions except superfluous content, the differences in rating were not statistically significant. All models had relevant difficulties considering the age-related aspects of the questions (AR 4.06-4.45, p=0.455). Conclusion This study highlights the potential and limitations of AI language models in addressing complex medical questions in fields characterized by highly individualized decision-making scenarios. The findings advocate for the development of more specialized AI tools in medicine, tailored to specific medical fields and patient age groups.Fig 1:Average ratings of LLMsFig 2:Rating differences between LLMs
Background: The interindividual variability of children's susceptibility to even slightly elevated IAP (intra-abdominal pressure) is unknown, as is the mechanism of transition from IAH (intra-abdominal hypertension = IAP > 10 mm Hg) to abdominal compartment syndrome (ACS = IAH+ organ dysfunction) and/or multiorgan dysfunction syndrome (MODS). Therefore, to improve the monitoring and assessability of micro- and macrocirculation in the context of IAP/IAH/ACS, we developed a multimodal monitoring concept and evaluated it as a prognostic tool.
Introduction: Liver transplantation (LT) in children has undergone significant changes over the last years. Especially the use of split and living donor transplants even for infants has led to new challenges in pre- and post-operative care. Systemic inflammatory response syndromes (SIRS, sepsis) as well-known complications after LT have not yet been systematically examined in the pediatric population. Methods: We analyzed clinical data of 39 pediatric liver transplant recipients regarding potential risk factors for post-transplant SIRS and sepsis. Secondly, the prognostic impact of SIRS and sepsis on post-transplant clinical course, patient and transplant-survival has been analyzed. Results: 64% of patients developed either SIRS (n = 16, 41%) or sepsis (n = 9, 23%) within 30 days after transplantation. No pre-transplant risk factors for increased susceptibility for SIRS or sepsis could be identified. Secondary closure of the abdomen (p = 0.045) and secondary biliary reconstruction (p = 0.043) were associated with a higher incidence of sepsis and were associated with significantly prolonged mechanical ventilation times in the presence of sepsis (p = 0.001). Patients with sepsis, but not SIRS, stayed significantly longer on PICU (p = 0.021) and suffered from higher mortality (n = 3 versus 0; p = 0.0006). All deaths within 30 days of transplantation were due to septic multiorgan failure. Neither early SIRS nor sepsis were associated with loss of transplant function. Conclusions: SIRS and sepsis are frequent events after pediatric liver transplantation. Sepsis increased length of PICU-stay and mortality significantly and prolonged duration of mechanical ventilation. Secondary biliary reconstruction and closure of the abdomen could be identified as potential risk factors for sepsis.
Sepsis in childhood is still a challenging diagnosis. Although the mortality of sepsis is much lower in children compared to adults, it remains one of the main causes of death in childhood. Increasing intensification of therapy in the field of neonatology and pediatric intensive care, the high proportion of patients with severe chronic pre-existing diseases in pediatric intensive care as well as the constantly growing number of children with inborn, acquired or therapy-induced incompetence of the immune system keeps the incidence of pediatric sepsis at a high level. In addition, there are an increasing number of cases of sepsis induced by highly resistant pathogens. One of the hallmarks of treatment success in pediatric sepsis is the early recognition of critically ill patients and the immediate initiation of effective treatment in order to significantly reduce the morbidity and mortality. To achieve these goals also in primary care settings, much effort is required in future years.
Die Sepsis stellt die Pädiatrie weiterhin vor große Herausforderungen. Obwohl die Mortalität der Sepsis im Kindesalter deutlich niedriger als bei erwachsenen Patienten ist, verursacht sie einen signifikanten Anteil der Sterberate bei Kindern. Die zunehmende Intensivierung der Therapie in der Neonatologie und der pädiatrischen Intensivmedizin, der höhere Anteil von Patienten mit schweren chronischen Vorerkrankungen in der Kinderintensivmedizin und die mittlerweile beträchtliche Anzahl von Kindern mit angeborener, erworbener oder therapeutisch induzierter Inkompetenz des Immunsystems halten die Sepsisinzidenz auf hohem Niveau. Hinzu kommt die steigende Zahl von Sepsen, die durch hochresistente Erreger induziert werden. Der gerade bei Kindern häufig fulminante Verlauf erfordert das sehr rasche Erkennen des kritisch kranken Sepsispatienten und den Beginn der sofortigen effektiven Therapie, um die Morbidität und die Mortalität signifikant zu senken. Hier sind noch große Anstrengungen notwendig, um die Therapie dahingehend wesentlich zu verbessern.
In pediatric patients with acute refractory cardiogenic shock (CS), extracorporeal membrane oxygenation (ECMO) remains an established procedure to maintain adequate organ perfusion. In this context, ECMO can be used as a bridging procedure to recovery, VAD or transplantation. While being supported by ECMO, most centers tend to keep their patients well sedated and supported by invasive ventilation. This may be associated with an increased risk of therapy-related morbidity and mortality. In order to optimize clinical management in pediatric patients with ECMO therapy, we report our strategy of veno-arterial ECMO (VA-ECMO) in extubated awake and conscious patients. We therefore present data of six of our patients with CS, who were treated by ECMO being awake without continuous analgosedation and invasive ventilation. Of these six patients, four were <1 year and two >14 years of age. Median time on ECMO was 17.4 days (range 6.9–94.2 days). Median time extubated, while receiving ECMO support was 9.5 days. Mean time extubated was 78 % of the total time on ECMO. Three patients reached full recovery of cardiac function on "Awake-VA-ECMO," whereas the other three were successfully bridged to destination therapy (VAD, heart transplantation, withdrawal). Four out of our six patients are still alive. Complications related to ECMO therapy (i.e., severe bleeding, site infection or dislocation of cannulas) were not observed. We conclude that "Awake-VA-ECMO" in extubated, spontaneously breathing conscious pediatric patients is feasible and safe for the treatment of acute CS and can be used as a "bridging therapy" to recovery, VAD implantation or transplantation.
Sepsis und septischer Schock sind Ursachen einer signifikanten Mortalität und Morbidität im Kindesalter. Frühzeitiges Erkennen des kritischen Zustands und sofortige aggressive Therapie sind für das Überleben und die Organprotektion entscheidend.
We report on an 11 years old patient that received living-related lung-lobar transplantation for severe respiratory insufficiency due to cystic fibrosis. His parents were suitable for donation, therefore his father donated the left lower lobe, his mother the right lower lobe.
In patients awaiting L u T x, MV and ECMO are often the last ways to create a bridge to L u T x. Both interventions are associated with a poor posttransplant outcome and survival rate. To improve the results of these patients, new “bridging‐strategies” are necessary. Recent reports demonstrate promising results for the concept of “awake ECMO ” in adult patients. To date, no data on this approach in pediatric patients have been available. We therefore describe the use of VV ‐ ECMO as a treatment strategy for RF in awake pediatric patients. It presents our experiences with the first three children treated using this new concept. Mean amount of time on ECMO was 44 days (range, 11.5–109 days). Two patients were successfully bridged to their L u T x. Both are still alive without any recurrences (24 and three months following L u T x). One patient died before a further L u T x after 109 days on ECMO due to adenoviral infection. Although reintubation was necessary in two patients, and total time being awake while on ECMO was <50%, we conclude that the concept of “awake VV ‐ ECMO ” is feasible for the treatment of RF and can be used as a “bridging therapy” to L u T x.
Objectives: In pediatric patients with cardiogenic shock (CS)/or acute cardiac failure (ACF) extracorporeal membrane oxygenation (ECMO) is often used as a bridging procedure to recovery or further implantation of a ventricular assist device (VAD). Although ECMO has been established as a routine procedure in pediatric tertiary care centers, the clinical path guiding these patients to fully recovery or further VAD implantation is one of the most challenging for pediatric intensivists.
Aim: Investigation of candidate SNPs previously described to influence the risk of developing childhood acute lymphoblastic leukemia, in pediatric non-Hodgkin lymphoma (NHL).
This single-centre, retrospective, observational pilot study was performed to evaluate the safety and efficacy of intravenous and oral itraconazole prophylaxis in paediatric haematopoietic stem cell transplantation (HCT). Study end-points were proven invasive fungal infection (IFI), survival, adverse reactions and graft-vs.-host disease (GVHD); 53 children and one young adult (median age 8.6 yr; range 0.4-18.3) transplanted between November 2001 and August 2004 were included in this study. Itraconazole was given intravenously from day +3 after HCT until oral medication became possible and continued until day +100 after HCT. Two proven new IFI in the itraconazole group (candidiasis, n = 1; aspergillosis, n = 1) were observed. After a median follow-up of 1.6 yr (0.3-6.1), six deaths (8%) were seen; 24 patients (45%) developed GVHD degree I-II, three children (6%) had GVHD degree III-IV. In 11 of 53 patients (21%), itraconazole prophylaxis was discontinued prematurely, mostly because of fever of unknown origin (n = 7). In total, 21 of 53 (40%) of the children had abnormal results of laboratory investigations during the prophylaxis. The results of this pilot study indicate that itraconazole prophylaxis during HCT in children is feasible and safe, despite abnormal laboratory results. The efficacy in terms of prevention of IFI, however, has to be addressed in a prospective large-scale study.
We show the effect of a network system in the treatment of pediatric septic shock, especially for children with Waterhouse–Friderichsen syndrome. In 2003 we founded a pediatric intensive care network with 15 children's hospitals in Lower Saxony, Germany. The aims were the standardisation of clinical therapies, implementation of training programs and the installation of an emergency system in the region of lower Saxony.
Hydrogen sulfide is produced endogenously by a variety of enzymes involved in cysteine metabolism.Clinical data indicate that endogenous levels of hydrogen sulfide are diminished in various forms of cardiovascular diseases.The aim of the current study was to investigate the effects of hydrogen sulfide supplementation on cardiac function during reperfusion in a clinically relevant experimental model of cardiopulmonary bypass.Twelve anesthetized dogs underwent hypothermic cardiopulmonary bypass.After 60 minutes of hypothermic cardiac arrest, reperfusion was started after application of either saline vehicle (control, n = 6), or the sodium sulfide infusion (1 mg/kg/hour, n = 6).Biventricular hemodynamic variables were measured by combined pressure-volume-conductance catheters.Coronary and pulmonary blood flow, vasodilator responses to acetylcholine and sodiumnitroprusside and pulmonary function were also determined.Administration of sodium sulfide led to a significantly better recovery of left and right ventricular systolic function (P < 0.05) after 60 minutes of reperfusion.Coronary blood flow was also significantly higher in the sodium sulfide-treated group (P < 0.05).Sodium sulfide treatment improved coronary blood flow, and preserved the acetylcholine-induced increases in coronary and pulmonary blood (P < 0.05).Myocardial ATP levels were markedly improved in the sulfide-treated group.Thus, supplementation of sulfide improves the recovery of myocardial and endothelial function and energetic status after hypothermic cardiac arrest during cardiopulmonary bypass.These beneficial effects occurred without any detectable adverse hemodynamic or cardiovascular effects of sulfide at the dose used in the current study.
PURPOSE:To analyze the association of genetic variation within the tumor necrosis factor (TNF -308 [G-->A]) and lymphotoxin alfa (LT-a +252 [A-->G]) genes with outcome in non-Hodgkin's lymphoma of childhood and adolescence.PATIENTS AND METHODS:Genotyping of the TNF -308 (G-->A) and LT-a +252 (A-->G) polymorphisms in patients (n = 488) enrolled onto the German-Austrian-Swiss multicenter trial NHL-BFM 95 from April 1996 to January 2000 was performed by polymerase chain reaction with subsequent restriction fragment length polymorphism analysis on DNA from tumor-free specimen.RESULTS:In patients with Burkitt's lymphoma (BL) and B-cell acute lymphoblastic leukemia (B-ALL; n = 219, 211 eligible patients), patients carrying at least two variant alleles (high-producer haplotypes) had an increased risk of events: probability of event-free survival (pEFS) at 3 years was 81% (SE = 5%), compared with 91% (SE = 2%) in low-producer haplotypes (P = .018). In BL/B-ALL with high tumor load (lactate dehydrogenase [LDH] > or = 500 U/L; n = 104), pEFS was 69% (SE = 8%) in high-producer versus 85% (SE = 4%) in low-producer haplotypes (P = .05). In multivariate analysis including risk factors for events (LDH > or = 500 U/L, CNS involvement, methotrexate infusion regimen), TNF -308/LT-alpha +252 haplotype kept prognostic relevance: patients with high-producer haplotypes had a 2.34-fold increase in risk of events (P = .048). The TNF -308 (G-->A) and LT-alpha +252 (A-->G) polymorphisms were not associated with pEFS in lymphoblastic lymphoma (n = 101), anaplastic large-cell lymphoma (n = 67), or diffuse large B-cell lymphoma (n = 65), nor with therapy-related toxicity.CONCLUSION:The TNF -308 (G-->A) and LT-a +252 (A-->G) polymorphisms were negative prognostic factors in pediatric BL/B-ALL. Among patients with serum LDH > or = 500 U/L, haplotype analysis further determined patients at risk for events.
Adenoviral (AdV) infections after transplantation remain a challenge in pediatric patients. Qualitative and quantitative PCR offer new approaches to early diagnosis and monitoring. However, their role in the management of AdV infections in pediatric transplant recipients remains to be determined. We report six children with positive qualitative serum-PCR for AdV on routine follow-up after transplantation (liver n = 4, hematopoetic stem cells (HSCT) n = 1, combined liver and HSCT n = 1). None of these children were symptomatic at the time of first detection of AdV. Two patients remained asymptomatic, one developed hemorrhagic cystitis and enteritis. Three children with positive PCR developed high viral load on quantitative PCR, all developed clinical AdV sepsis with further rising virus load. Despite antiviral therapy with cidofovir, these three patients died of septic multiorgan failure. Positive qualitative AdV-PCR from blood after pediatric transplantation is not necessarily followed by clinical disease. In case of positive AdV-PCR, monitoring by serial quantitative PCR is useful regarding treatment decision and prevention of fatal disease.
In the Non-Hodgkin Lymphoma-Berlin-Frankfurt-Münster 95 (NHL-BFM95) study, we tested by randomization whether for patients with B-cell neoplasms methotrexate as intravenous infusion over 4 hours (MTX-4h) is not inferior to, but less toxic than, a 24-hour intravenous infusion (MTX-24h). Second, we investigated against the historical control of study NHL-BFM90, whether for patients with moderate tumor mass MTX can be reduced from 5 g/m(2) to 1 g/m(2). Patients received 2 5-day therapy courses in risk group R1 (resected), 4 in R2 (lactate dehydrogenase [LDH] < 500 U/L), 5 in R3 (LDH > 500 to < 1000 U/L) and 6 in R4 (LDH > 1000 U/L and/or central nervous system [CNS] disease). Courses contained MTX 1 g/m(2) in R1 + R2 and 5 g/m(2) in R3 + R4. Of 505 patients (April 1996 to March 2001), 364 were randomized to receive MTX-4h or MTX-24h. Failure-free survival (pFFS, 1 year) for arm MTX-4h versus MTX-24h, respectively, was 95% +/- 5% (n = 20) versus 100% (n = 19) in R1, 94% +/- 2% (n = 88) versus 96% +/- 2% (n = 95) in R2, and 77% +/- 5% (n = 62) versus 93% +/- 3% (n = 69) in R3 +/- R4 (per-protocol analysis). Incidence of mucositis grade III/IV was significantly lower with MTX-4h in all risk groups. For patients in R2, event-free survival (pEFS) was 95% +/- 2% (n = 222) in NHL-BFM95 (MTX 1 g/m(2)) and 97% +/- 1% (n = 154) in NHL-BFM90 (MTX 5 g/m(2)). In conclusion, MTX-4h was less toxic than MTX-24h. MTX-4h was noninferior to MTX-24h for limited stage B-cell non-Hodgkin lymphoma (B-NHL) but not for advanced disease. For limited disease, MTX 1 g/m(2) is noninferior to 5 g/m(2).