Preliminary exploration of tissue oxygen saturation index using near-infrared spectroscopy in two infants of neonatal catheter-associated venous thrombosis

Abstract

Background

Near-infrared spectroscopy (NIRS) is commonly used for monitoring oxygen saturation in brain and abdominal tissues, while reports on the monitoring of local tissue oxygen saturation in limbs are very limited. This study observed the changes in limb tissue oxygen saturation index (TOI) in two infants of catheter-related venous thrombosis (CRT) and preliminarily investigated their correlation.

Case presentation

This study involved two infants who experienced limb catheterization complicated by venous thrombosis. NIRS was employed for the continuous monitoring of TOI prior to, during, and subsequent to urokinase thrombolysis. The safe TOI range was established at 58%–68% ±2; a reduction above 20% from baseline signified an anomaly. Changes in skin color and temperature of the affected limb were simultaneously observed, and the thrombolytic effect was comprehensively assessed using vascular ultrasound. Infant 1 was diagnosed with right iliac vein thrombosis; after 144 h of thrombolytic therapy, the TOI of the affected lower limb returned to the normal range, with no abnormalities in limb skin temperature or circumference. Infant 2 involved axillary vein thrombosis near the subclavian vein; following 17 h of combined thrombolysis and anticoagulation therapy, TOI values in both upper limbs became balanced, and ultrasound follow-up revealed no residual thrombus, indicating restored venous patency.

Conclusions

NIRS-monitored limb tissue oxygen saturation index can indicate the local venous filling status, which may facilitate the assessment of thrombolytic efficacy and the surveillance of thrombus formation.

Introduction

The regional tissue oxygen saturation or tissue oxygen saturation index (TOI) is measured by near-infrared spectroscopy (NIRS), which also determines the whole body hemodynamic status and the perfusion status of various tissues and organs. NIRS is noninvasive, suitable for bedside use, and real-time data enables continuous monitoring [1]. NIRS typically employs a wavelength range of 700–1000 nm for near-infrared radiation. This band provides the ability to monitor oxygen saturation levels in the brain, kidneys, intestines, and muscle by penetrating bone, epidermis, fat, and muscle [2]. Nevertheless, prior investigations of the utilization of NIRS have typically concentrated on the monitoring of oxygen saturation in the brain and abdominal tissues, rather than the local tissues of the extremities [3]. The present case reports investigated the changes in limb tissue perfusion and blood oxygen saturation index before and after thrombolytic therapy in two infants with venous thrombosis resulting from catheterization. This was accomplished by monitoring the limb TOI value. The readings from the thrombosis-free limb on the opposing side were evaluated as controls.

Case presentation

Infant one was male, 20 days old, with a birth weight of 1940 g, and was admitted on September 8, 2022. On the second day following admission, peripherally inserted central catheter (PICC) catheterization was performed through the right superficial temporal vein, and 12 cm was inserted and 4 cm was exposed. On the fourth day, laparotomy plus intestinal adhesion release, intestinal resection and anastomosis, appendectomy, and abdominal irrigation and drainage were performed under general anesthesia. Ultrasound-guided right femoral vein puncture and indwelling central venous catheter (CVC) tube were performed in the department of anesthesiology before the operation, and the patients were transferred to NICU for monitoring after operation. The body temperature was 35℃, pulse was 164 beats/min, respiration was 34 beats/min, blood pressure was 80/53 (62) mmHg, and oxygen saturation was 94%. The skin of the right lower limb was whiter than that of the left lower limb, and the extremities were colder. The CVC catheter of the right femoral vein was removed for 3 hours. The skin of the right lower limb was still purple and black, and the extremities were cold. The bilateral femoral circumference was 12 cm. Three hours after the removal, a vascular ultrasound examination showed right iliac vein thrombosis, 1.2 cm * 0.26 cm, color Doppler flow imaging (CDFI): local filling defect in the vein; the TOI value was 45.7% in the right lower limb and 58.7% in the left lower limb. The right lower limb was lower than the left lower limb. And the relative percentage decrease is 22.15%. Urokinase 4400 u / kg / IV was given through the PICC catheter, then 1 ml / h = 4400 u / kg / hr continuous infusion. After 68 hours of thrombolysis, the right external iliac vein was detected by B-ultrasound as 0.69 cm * 0.16 cm, with CDFI: a local filling defect in the vein. The TOI value of the right lower limb was 55.5%; the left lower limb was 61.4%. The right lower limb was still lower than the left lower limb. And the relative percentage decrease is 9.61%; the skin of the limb was relieved, and the extremities were warm. After 144 hours of application, B-ultrasound was used to examine the local vascular roughness of the right external iliac vein. The TOI value was 59.2% in the right lower limb and 59.8% in the left lower limb. The previous skin abnormalities of the right lower limb were resolved. The extremities were warm, and the bilateral femoral circumference was 11.5 cm. Prothrombin: prothrombin time (PT) 11.9/13.6”, activated partial thromboplastin time (APTT) 27.7/52.9”, clotting time (TT) 19.6/20.6”, and plasma D-dimer determination 8.52 mg/L (up to 17.20). PICC was retained for 379 hours.

Infant two was a female, 6 days old, G1P1, 39 weeks of pregnancy, cesarean section, birth weight 2900 g, admitted on 2022.08.21. On admission, the body temperature was 36.9°C, the pulse was 140 beats/min, the respiration was 46 beats/min, and the blood pressure was 55/32 mmHg. On the 23rd day of admission, the left basilic vein PICC catheter was placed. On the 5th day of catheterization, the left upper arm was measured to be 0.5 cm thicker than the right upper arm. The blood returned smoothly, and the use was suspended. The next day, B-ultrasound consultation and vascular ultrasound examination: axillary vein near the subclavian vein thrombosis (2.7 cm * 0.2) adhered to the lateral wall of the catheter; CDFI: axillary vein local blood flow filling defect; arm circumference left 11 cm, right 10 cm; the TOI value was 43.9% in the left upper limb and 61.9% in the right upper limb, and the relative percentage decrease is 29.08%. The left upper arm was 0.5–1.0 cm thicker than the right upper arm. Urokinase 4400 u/kg/IV after 1 ml/h = 4400 u/kg/hr continuous pump injection, application of 4 hours of Nadroparin Calcium 0.035 ml subcutaneous injection q12h * 2 times. A B-ultrasound examination of the left axillary vein near the subclavian vein at 10 hours after thrombolysis detected a thrombus of 1.65 cm * 0.12 cm. CDFI: local filling defect in the vein. The TOI value was 50.6% in the left upper limb and 59.8% in the right upper limb. The left upper limb was still lower than the right upper limb. And the relative percentage decrease is 15.38%. The left upper arm was still thicker than the right upper arm by 1.0 cm. After 17 hours of thrombolysis and anticoagulation therapy, B-ultrasound showed no obvious thrombosis in the left axillary vein and the left subclavian vein. The CDFI demonstrated local blood flow within the axillary vein. The TOI value was 59.5% in the left upper limb and 59.3% in the right upper limb. There was no difference in bilateral arm circumference. Prothrombin: prothrombin time (PT) 13.5”, activated partial thromboplastin time (APTT) 54.7”, clotting time (TT) 13.3”, and plasma D-dimer determination 1.78 mg/L. The PICC was maintained for 137 hours. The clinical data of 2 CRT infants are shown in Tables 1 and Table 2. The clinical outcomes of the infants before and after treatment and thrombolysis are shown in Tables 3, 4, and 5.

Table 1.

Clinical data of 2 infants with CRT

Case number	Age of diagnosed CRT (d)	Catheter type	Time of catheter placement about diagnosis of CRT (h)	Catheter removal time after diagnosis of CRT (h)	Tube placement hours (h)	High risk factors	Thrombus site	Thrombus size	Thrombosis related symptoms	Thrombolytic/anticoagulant drugs	Course of thrombosis treatment (h)	Thrombus disappeared until CRT diagnosis time (h)
1	20	CVC	4	3	3	premature infants, necrotizing enterocolitis, exploratory laparotomy	Right iliac vein	1.2 cm*0.26 cm	The skin on the right lower limb changed color to purple-black and felt cool.	urokinase	144	160
2	6	PICC	114	23	137	Congenital heart disease, pulmonary hypertension, neonatal necrotizing enterocolitis, neonatal sepsis	Subclavian vein	2.7 cm*0.2 cm	The left upper arm is 0.5 cm thicker than the right upper arm	Urokinase/nardraparin calcium	17	23

CRT Catheter related venous thrombosis

Table 2.

Comparison of coagulation parameters before and after thrombolytic therapy in two CRT infants

Time	PT(s)	APTT(s)	TT(s)	D-dimer(mg/L)
Before treatment	14/12.5	63.6/54.7	17.6/13.3	7.48/1.78
During treatment	13.9/-	59.6/-	18.9/-	17.2/-
After treatment	13.6/-	52.9/-	20.6/-	8.52/-

Infant 1 / Infant 2; PT Prothrombin Time, APTT Activated Partial Thromboplastin Time, TT Thrombin Time

Table 3.

Comparison of TOI values between affected and contralateral limb before and after thrombolytic therapy in infants

	Before treatment		After treatment
	Affected limb(%)	Contralateral limb(%)	Affected limb(%)	Contralateral limb(%)
Infant 1	45.7	58.7	59.2	59.8
Infant 2	43.9	61.9	59.5	59.3

Table 4.

Comparison of limb circumference between affected and contralateral limb before and after thrombolytic therapy in infants

	Before treatment		After treatment
	Affected limb(cm)	Contralateral limb(cm)	Affected limb(cm)	Contralateral limb(cm)
Infant 1	12	12	11.5	11.5
Infant 2	11	10	10	10

Table 5.

Comparison of thrombus size and CDFI between affected and contralateral limb before and after thrombolytic therapy in infants

	Before treatment		After treatment
	Thrombus size(cm)	CDFI	Thrombus size(cm)	CDFI
Infant 1	1.2*0.26	local filling defect in the vein	0.69*0.16	axillary vein local blood flow filling defect
Infant 2	2.7*0.2	local filling defect in the vein	Not obvious	local blood flow within the axillary vein

Nursing care of children with CRT

TOI monitoring method

The EGOS-600 A near-infrared spectrometer, produced by Suzhou Aiqin Biomedical Electronics Co., Ltd., was utilized, with a measurement depth of 0.5–2.5 cm. The emission wavelengths of the near-infrared light were 760 nm, 810 nm, and 840 nm. The data safety range was defined as 58–68% ± 2 or a relative decrease exceeding 20% from baseline values. The instrument collected data every 2 s and employed spatial resolution algorithms to calculate estimated concentrations of oxyhemoglobin (CHbO₂) and deoxyhemoglobin (CHb). The TOI value was determined as the ratio of CHbO₂ to total hemoglobin (CtHb = CHbO₂+ CHb) [4, 5]. Subsequent to monitoring, the data—stored in bin file format by the NIRS instrument—was exported and transformed into Excel files using the associated software DateView.exe, after which the average TOI value was computed. The B-type probe was employed for monitoring purposes. The maximum duration of probe application in a single location on the human body was limited to 12 h for newborns and 4 h for preterm infants to prevent skin compression injury. In two infants exhibiting thrombosis symptoms, vascular B-ultrasound confirmed the diagnosis of thrombosis. The NIRS probe was positioned at the flat distal end of the affected limb and similarly on the contralateral leg to monitor the TOI values of both limbs. The monitoring site of infant 1 was the lower limbs, and the probe was pasted on the fibular flat of both lower limbs, and the pasting site was changed every 4 h. In newborn 2, the monitoring site was the upper limb, with the probe affixed to the palmar aspect of both upper arms, and the attachment point was altered throughout each shift. Throughout the procedure, the child was in a quiet state, the instrument probe and the skin were well fitted, and 3 M tape was used to prevent light leakage. The skin at the monitoring site was intact without compression symptoms. In the ongoing assessment of TOI, the TOI value recorded from the moment of B-ultrasound diagnosis to the initiation of treatment was designated as the pre-thrombolysis TOI value. The resolution of thrombus and limb symptoms identified via B-ultrasound constituted the TOI value following thrombolysis. The results obtained from the intermediate B-ultrasound examination were utilized as the TOI value during thrombolysis. Figure 1 illustrates this.

Fig. 1.

B-ultrasound results before, during and after thrombolysis with tissue oxygen saturation index (TOI) monitoring values. A Case 1-A1 before thrombolysis. B Case 1-A2 during thrombolysis. C Case 1-A3 after thrombolysis. D Case 1-B1 before thrombolysis. E Case 1-B2 during thrombolysis. F Case 1-B3 after thrombolysis. G Case 2-A1 before thrombolysis. H Case 2-A2 during thrombolysis. I Case 2-A3 after thrombolysis. J Case 2-B1 before thrombolysis. K Case 2-B2 during thrombolysis. L Case 2-B3 after thrombolysis

CRT thrombolytic therapy regimen

The Dutch national guidelines for newborn critical care transport prescribe the following on the management of central venous catheter-related thrombosis [6]: In cases of asymptomatic thrombosis, a “wait-and-see” approach may be implemented; for symptomatic thrombosis, anticoagulation therapy is advised; in instances of high-risk thrombosis, thrombolytic therapy followed by anticoagulation is recommended, with high-risk venous thrombosis defined as thrombi that pose a threat to a limb or organ. Due to the significant risk of treatment-related hemorrhage, thrombolytic therapy is contraindicated until thromboembolism has resulted in substantial organ or limb injury [7]. Urokinase can be administered as an initial bolus of 4,400 IU/kg over 20 min, followed by a continuous infusion of 4,400 IU/kg per hour for 6 to 12 h. The treatment should be discontinued when ultrasound confirms the disappearance of the thrombus [8]. In infant 1, the skin of the right lower limb turned from pale to purplish-black, with cold extremities; in infant 2, the left upper arm was 0.5 cm thicker than the right upper arm. Both cases received thrombolytic therapy with urokinase: an intravenous bolus of 4,400 IU/kg was administered over 5–10 min, followed by continuous infusion at a rate of 4,400 IU/kg per hour. Infant 1 continued the infusion for 144 h, while infant 2 switched to nadroparin calcium anticoagulation after 4 h of continuous infusion. Before thrombolysis, the APTT of both children was significantly elevated by approximately 1.67–1.44 folds (ranging from 63.6 s to 54.7 s). At the final stable state, the APTT of infant 1 was 59.6 s with no significant fluctuations, whereas APTT monitoring was discontinued in infant 2. During the treatment period, both children were closely monitored and showed no bleeding tendency.

Discussion

Feasibility of monitoring TOI using NIRS

The objectives of the treatment and monitoring of catheter-related venous thrombosis are to prevent thrombus propagation and progression, restore blood perfusion to the affected organs, and minimize long-term adverse outcomes [9]. The entire process of venous thrombosis, from initial onset, extension, and expansion to organization and recanalization, involves damage to venous valves and intima and is associated with notable complexity. Due to the small venous lumen of the newborn and the slow blood flow rate after thrombosis, it is prone to poor venous filling. However, the detection rate of thrombotic events using ultrasound ranges from 14% to 35%, which may lead to misdiagnosis or missed diagnoses [10]. A study monitored 80 pediatric patients with limb catheterization (including 27 newborns with femoral vein/lower limb PICC) and found that the lower limb tissue oxygen saturation in the thrombosis group was significantly lower than on the healthy side. NIRS can serve as an early warning for venous stasis before limb swelling or cyanosis occurs and is recommended as a routine screening method for lower limb catheterization in newborns [11]. The study compared oxygen consumption calculated by NIRS with techniques such as Doppler ultrasound (measuring blood flow in major arteries) and plethysmography (assessing whole-body tissue blood flow), confirming the reliability of NIRS measurements [12–13]. The difference lies in the fact that other monitoring methods primarily reflect overall blood flow, whereas NIRS focuses on local blood flow, making changes in NIRS parameters more pronounced during exercise [14]. By continuously and in real-time monitoring limb TOI values with NIRS, CtHb mainly reflects local tissue perfusion status. Nurses can perform this bedside operation independently and, by combining clinical symptoms, promptly identify catheter thrombosis and initiate thrombolytic therapy early.

The correlation between TOI value and CRT formation and thrombolysis effect

Studies have shown that when muscle tissue perfusion decreases by approximately 50% and at lower blood flow levels, the TOI value exhibits an almost linear relationship with muscle tissue perfusion [15]. When the thrombus was diagnosed by color Doppler ultrasound, and the local filling defect in the vein was visualized by CDFI, the TOI value of the thrombus side limb was (45.7%, 43.9%), which was lower than the low value of the safe range (56%). The TOI values in the contralateral limbs were (58.7%, 61.9%), and the relative percentage decrease are 22.15% and 29.08%. B-ultrasound examination showed that the thrombus was reduced, the skin color and temperature of the extremities were significantly improved at 68 h after thrombolysis in infant 1 and 10 h after thrombolysis in infant 2, but CDFI still showed local filling defects in the veins. The TOI values of the thrombosed limb (55.5%, 50.6%) were below the lower limit of the safe range, but both were higher than those of the contralateral limb (61.4%, 59.8%). And the relative percentage decrease is 9.61% and 15.38%. When the B-ultrasound examination showed that the thrombus disappeared, CDFI showed local filling in the vein, and the symptoms related to limb thrombosis disappeared, the measured TOI values of the limbs were within the safe range. Therefore, in practical application, the thrombolysis effect can be judged according to whether the TOI value is in the safe range and the difference of the basic value of 20%. Moreover, neonatal limb-specific baseline values can vary, making the contralateral limb an essential internal control for such monitoring.

Conclusions

NIRS enables non-invasive, real-time, and continuous monitoring of tissue oxygenation. Specifically, the limb tissue oxygen saturation index reflects local venous perfusion and filling status—particularly in the microcirculatory bed—making it a physiologically plausible surrogate for venous congestion. In clinical settings where bedside Doppler ultrasound is delayed, inaccessible, or limited by operator experience, serial NIRS-derived limb tissue oxygen saturation index measurements may offer a complementary, objective, and dynamic parameter to support early suspicion of deep vein thrombosis formation and to monitor response to thrombolytic therapy. The current study included a limited number of cases. And further accumulation of cases with increased sample size is needed to better summarize changes in limb tissue oxygen saturation index following CRT formation, explore their associations, and guide thrombolytic and anticoagulant therapy for symptomatic CRT in newborns, thereby reducing the occurrence of related complications.

Abbreviations

NIRS

Near-infrared spectroscopy

TOI

Tissue oxygen saturation index

TPN

Total parenteral nutrition

PICC

Peripherally inserted central catheter

CVC

Central venous catheter

CDFI

Color doppler flow imaging

NEC

Neonatal necrotizing enterocolitis

NICU

Neonatal intensive care unit

PT

Prothrombin time

APTT

Activated partial thromboplastin time

TT

Clotting time

CHbO₂

Concentrations of oxyhemoglobin

CHb

Deoxyhemoglobin

CtHb

Total hemoglobin

CRT

Catheter-related venous thrombosis

Authors’ contributions

Haihong Zhu: Writing – original draft, Supervision, Validation, Funding acquisition. Yun Ling: Writing – review and editing, Conceptualization, Investigation. Xiaoying Dong: Writing – review and editing, Visualization, Investigation. Feixiang Luo: Writing – review and editing, Formal analysis, Project administration. Jihua Zhu: Writing – review and editing, Supervision, Methodology.

Funding

This work is supported by Zhejiang Provincial Medical and Health Science and Technology Program: 2025KY937.

Data availability

All data utilized in this work can be obtained from the corresponding author upon request.

Written informed consent for publication of clinical details and/or clinical images had been obtained from the parents of the patient.

Declarations

Ethical approval and consent to participate

The ethical approval was obtained from the Ethics Committee of Children’s Hospital, Zhejiang University School of Medicine (2017-IRB-015). All study participants or their guardians provided written informed consent before enrollment.

Consent for publication

Written informed consent for publication of clinical details and clinical images was obtained from the parents of patient.

Competing interests

The authors declare no competing interests.