Introduction

Intravenous chemotherapy is a cornerstone treatment in modern medicine. However, administration of vesicants1 or irritants2 often leads to a local venous inflammation with manifestations including erythema, pain, edema, increased warmth, palpable cord, and purulent drainage, namely chemical phlebitis3,4. These symptoms may persist for months, which profoundly impact patients’ quality of life and normal daily activities5. Despite advances in preventive strategies6, chemical phlebitis remains a prevalent complication, particularly among pediatric7 and malignant tumor patients8. Therefore, it is necessary to explore the pathophysiological features of chemical phlebitis using experimental animal models, which may help raise awareness of its harm.

Current animal models of chemical phlebitis primarily rely on intravascular injection, for rabbits via the marginal ear vein9,10,11,12,13 and mice via the lateral tail vein14,15, which is more for the former. However, these models present several anatomical and practical limitations. The rabbit auricle is composed primarily of thin skin and cartilage tissue, with the marginal ear vein being superficial yet highly branched, featuring numerous collateral vessels. Similarly, the mouse tail is covered by circular keratin scales and dense collagen, where the lateral tail vein is both fine and deeply positioned, rendering intravenous injections technically challenging. Crucially, both the auricle and tail lack substantial soft tissue components such as muscle and adipose tissue. As a result, animal models utilizing these sites fail to fully replicate the clinical manifestations of chemical phlebitis. Additionally, it is difficult to sample lesioned vein specimens from these sites for histological studies, particularly for examining tunica intima microstructure and endothelial cell morphology.

In this study, a rat model of chemical phlebitis was established by intravascular administration of vinorelbine solution in a concentration-gradient manner via the dorsalis pedis vein of the hindpaw. The severity of phlebitis was evaluated using a standardized phlebitis scale, which assessed clinical indicators including erythema, pain, edema, palpable cord formation, and purulent drainage. Furthermore, pathological featues of venous tissues was observed by hematoxylin-eosin and masson’s trichrome staining, as well as scanning and transmission electron microscopy.

Materials and methods

Animals

A total of 48 male Sprague-Dawley rats (9 weeks old, 300 ~ 320 g) were acquired from Beijing HFK Bioscience Co., Ltd (China) and housed in a temperature-controlled room (22 ~ 24 °C) with a 12/12 light/dark cycle and fed with standard rat chow and tap water. All procedures were approved by the Medical Ethics Committee of the Shanxi Academy of Traditional Chinese Medicine (Approval Number: SZYLY2022KY-0302). All experiments were performed in accordance with relevant guidelines and regulations, and the authors complied with the ARRIVE guidelines.

Modeling method

The rats were randomly divided into 6 groups, with each 8 in each group. The right hindlimb hair was removed by depilatory cream 1 day earlier. Then the rat was anesthetized with 2% isoflurane and the right knee was wrapped by a rubber tourniquet to engorge the dorsalis pedis vein for easy puncture. The chemical phlebitis was induced in rats of 5 groups by injecting 0.1 mL of 1 mg/ mL, 2 mg/mL, 3 mg/mL, 4 mg/mL, and 5 mg/mL vinorelbine solution, respectively. While the remaining rats were received 0.1 mL of normal saline serving as sham group. The puncture site was pressed for 5 min to prevent bleeding or drug leakage after injection. Then the general symptoms of phlebitis were observed daily. On day 7, the animals were euthanized with Euthasol (100 mg/kg, IV), and a 1-cm segment of the vein extending proximally from the puncture site in the right hindlimb was collected for histopathological examination.

Hindlimb volume test

The volume of the injected hindlimb (2 cm above the ankle) in rat was measured by drainage method16 before modeling and post-model daily for 7 days. The edema was quantified as swelling rate, which is calculated by the following formula:

$${\text{Swelling~rate~(\% )~ = ~}}\frac{{{\text{Post - model~volume~}} - {\text{~initial~volume}}}}{{{\text{Initial~volume}}}} \times {\text{100\% }}$$

Bipedal balance test

Pain was evaluated by weight-bearing ratio of treated hindlimb, which is determined by the bipedal balance test17 before modeling and post-model daily for 7 days. Briefly, the rats were placed in the bipedal balance analgesiometer and make sure both of the hind paws stand on the corresponding pressure sensors for recording the bilateral paw pressures, respectively. The weight-bearing ratio is calculated by the following formula:

$${\text{Weight - bearing~ratio~(\% ) = ~}}\frac{{{\text{Pressure~on~the~injected~paw}}}}{{{\text{Total~pressure~of~both~paws}}}} \times {\text{100\% }}$$

Phlebitis grading

The symptoms of chemical phlebitis including erythema, pain, edema, palpable cord, and purulent drainage in the injected hindlimb were observed each day for 7 and the severity was graded according to the phlebitis scale18 shown in Table 1.

Table 1 Phlebitis grading scale.
Full size table

Hematoxylin-eosin and masson’s trichrome staining

The venous tissues were fixed in 4% polyformaldehyde, dehydrated and embedded in paraffin blocks. Subsequently, transverse sections of 5 μm were stained with hematoxylin-eosin (HE) and masson’s trichrome staining, respectively. Images were visualized using an M8 digital scanning microscopy imaging system (Lecia).

Electron microscopy

The venous tissues were pre-fixed with 3% glutaraldehyde at 4 °C for 2 days and post-fixed with 1% osmic acid for 2 h. Subsequently, the tissues were gradually dehydrated with 50–100% ethanol. For scanning electron microscopy (SEM), the longitudinal section of veins was treated by isoamyl acetate for 15 min and dried by a critical point dryer. Specimens were attached to metallic stubs by carbon stickers and sputter-coated with gold for 30 s. Then the sections were observed and captured by SEM (HITACH, SU8100). For transmission electron microscopy (TEM), transverse sections of veins were immered in 100% acetone/Epon 812 and ultrathin slices (60 nm) were prepared. The sections were stained with 5% uranium acetate for 30–60 min and then with lead citrate for 10 min. After drying, the sections were observed and captured by TEM (HITACH, HT7800).

Statistical analysis

The data were analyzed using the SPSS software version 22.0. The swelling rate and weight-bearing ratio were presented as mean ± standard deviation. Comparisons between the two groups were conducted using the independent t-test. The Kruskal-Wallis one-way analysis of variance was used to analyze the grading of phlebitis. A p-value of less than 0.05 was considered statistically significant.

Results

General symptoms in hindlimbs

The general symptoms in hindlimb of rats following intravenous injection of saline or vinorelbine via the dorsal pedal vein were observed daily for 7 days. Representative photographs were captured every other day, as depicted in Fig. 1. During observation period, sham-operated rats (0 mg/mL) exhibited no signs of erythema or edema in the hindlimb with skin intact. In contrast, rats administered vinorelbine displayed marked erythema and edema as early as day 1 post-injection, with symptom severity escalating in a dose-dependent manner. By day 3, these manifestations progressed, accompanied by ulcer formation. Subsequently, partial alleviation of erythema and edema was observed by day 5. However, these improvements were followed by the development of induration and streak formation by day 7.

Fig. 1
Fig. 1
Full size image

The general symptoms of injected hindlimb in rats induced by different concentrations of vinorelbine via dorsal pedal vein, with images taken every other day.

Edema evaluation

As illustrated in Table 2, the swelling rate of hindlimb in rats treated by all concentration of vinorelbine significantly increased compared to the sham rats on the 1st day post-injection. Furthermore, the swelling rate increased in accordance with the ascending concentration of vinorelbine solution and the peak was observed on the 3rd day. After that, the swelling rate exhibited a slight decrease on the 5th day but still remained above 30% in the 3 mg/mL, 4 mg/mL and 5 mg/mL vinorelbine groups on the 7th day.

Table 2 Comparison of swelling rates induced by different concentration of vinorelbine.
Full size table

Pain evaluation

Pain was identified when the weight-bearing ratio of injected paw in bipedal stance testing fell below 50%. As shown in Table 3, the weight-bearing ratio of injected paw in sham group maintained about 50% during the experimental period. In contrast, the ratio in rats injected with vinorelbine solution may less than 50%, the extent of which depended on the timepoint and vinorelbine concentration. On the 1st day, the ratio in rats of all vinorelbine-treated groups with a slight decrease tendecy was observed compared to the sham rats. Obviously, the ratio was decreased markly on the 3rd and 5th day, especially in the 3 mg/mL, 4 mg/mL and 5 mg/mL vinorelbine groups. By the end of day 7, the ratio of injected paw in rats treated by 4 mg/mL and 5 mg/mL vinorelbine still significantly decreased compared to the sham rats.

Table 3 The weight-bearing ratio of the injected hindlimb in rats.
Full size table

Grading of chemical phlebitis by phlebitis scale

The chemical phlebitis in each rat was graded by phlebitis scale on day 7 and the results were shown in Table 4. The severity of chemical phlebitis among these groups was correlated with the concentration of vinorelbine solution. All the rats in sham group (0 mg/mL) and 1 mg/mL vinorelbine group were graded 0 with no symptoms. While in the 2 mg/mL vinorelbine group, 37.5% of the rats were graded 0 and 62.5% displayed grade I. The incidence of grade I, II and III in the 3 mg/mL group were 25%, 50% and 25%, respectively. In the 4 mg/mL group, the incidence of grade II, III, and IV were 25%, 50% and 25%, respectively. The grade III and IV account for 37.5% and 62.5% in the 5 mg/mL group, respectively.

Table 4 The grading of chemical phlebitis assessed by phlebitis scale on day 7.
Full size table

Hispathological features in venous tissues

Hematoxylin-eosin staining was performed on venous tissues sampled from the injected hindlimb of rats on day 7 with representative graphs shown in Fig. 2. Venous tissues from 0 mg/mL exhibited normal morphology, characterized by flexible vessel walls with classical layers of tunica adventitia, media and intima. The intima displayed hobnailed endothelium and the lumen was free of thrombus. Veins from rats treated with 1 mg/mL vinorelbine showed nearly identical histological features to normal veins, with no significant pathological alterations. However, the thicken vein wall were presented in the venous tissues collected from the rats injected 2 mg/mL vinorelbine. More pronounced pathological changes were observed in the 3, 4, and 5 mg/mL vinorelbine groups. These veins exhibited marked wall thickening, luminal stenosis, and thrombi within the circular venous lumen, indicating progressive dose-dependent venous damage.

Fig. 2
Fig. 2
Full size image

Representative histopathology images of venous tissues collected from rat hindlimbs stained by hematoxylin-eosin. above, 80×magnification, scale bar = 100 μm; below, 400×magnification, scale bar = 20 μm.

Furthermore, venous tissue morphology was assessed using Masson’s trichrome staining, as shown in Fig. 3. In venous tissues from the 0 mg/mL and 1 mg/mL vinorelbine groups, endothelial cells rested on the subendothelial connective tissue. Additionally, an internal elastic lamina, composed of aggregated elastic fibers, separated the tunica intima from the tunica media. In contrast, veins injected with higher concentrations of vinorelbine exhibited dose-dependent pathological changes, including wall thickening, luminal stenosis and thrombus. The thickened vein wall with intimal hyperplasia was prominent in the 3, 4, and 5 mg/mL vinorelbine groups, which may result in reduced deformabilit. The neointima was primarily composed of smooth muscle cells interspersed with collagen fibers, while abundant collagen deposition was observed in the tunica adventitia.

Fig. 3
Fig. 3
Full size image

Representative histopathology images of venous tissues collected from rat hindlimbs stained by masson’s trichrome. Collagen fibers were stained blue, muscle fibers and erythrocytes were stained red. above, 80×magnification, scale bar = 100 μm; below, 400×magnification, scale bar = 20 μm.

Microstructural features in venous tissues

On day 7 post-injection, longitudinal venous tissues from rat hindlimbs were examined by scanning electron microscopy as shown in Fig. 4. In both 0 mg/mL and 1 mg/mL vinorelbine-treated rats, venous tunica intima exhibited a smooth, continuous surface composed of polygonal endothelial cells aligned parallel to the vascular longitudinal axis. Notably, local tunica intima in the 2 mg/mL vinorelbine group adhered to a substantial number of activated platelets. At higher concentrations (3–5 mg/mL), more pronounced dose-dependent mirstructural alterations of lumen surface was observed, characterized by rough with blood cells, densely packed spindle-shaped endothelial cells, and significant accumulation of extracellular matrix.

Fig. 4
Fig. 4
Full size image

Representative microstructure images of vascular intima observed by scanning electron microscopy. above, 500×magnification, scale bar = 100 μm; below, 3000×magnification, scale bar = 10 μm.

Further observation of venous transverse sections by transmission electron microscopy as shown in Fig. 5 revealed distinct morphological changes across treatment groups. Venous tissues from 0 mg/mL and low-dose vinorelbine-treated (1–2 mg/mL) rats maintained normal vascular wall architecture, characterized by a tightly aligned monolayer of endothelial cells with prominent nuclei along the luminal surface and an intact basement membrane. In contrast, venous tissues from rats treated with higher vinorelbine concentrations (3 mg/mL, 4 mg/mL, and 5 mg/mL) displayed significant vascular wall thickening. The luminal surface in these groups exhibited crowded arrangements of shrunken endothelial cells intermixed with extracellular matrix.

Fig. 5
Fig. 5
Full size image

Representative microstructure images of vascular wall observed by transmission electron microscopy. above, 200×magnification, scale bar = 50 μm; below, 2500×magnification, scale bar = 5 μm.

Discussion

Vinorelbine is a semi-synthetic vinca-alkaloid with a broad spectrum of anti-tumor activity, which has been approved for treating non-small cell lung cancer, breast cancer, ovarian cancer, etc19. The recommended dosage for human is 25–30 mg/m2 and hematological toxicity such as leucopenia is one of dose-limiting adverse reactions. Additionally, the incidence of phlebitis after vinorelbine injection from the peripheral superffcial vein can be as high as 30%20. The median lethal dose (LD50) for intravenous administration in rats is 11.2 mg/kg. In our preliminary studies, leucopenia was observed in rats intravenous infused a single dose of 6 mg/kg vinorelbine within 1 day and lasted for at least 7 days. Meanwhile, a significant weight loss also occurred in these rats. In contrast, there were almost no such abnormal events in rats subjected to 3 mg/kg by the same way. So, we designed a protocol of 0.1 mL vinorelbine with a concentration-ascending mode via the dorsalis pedis vein to induce phlebitis in rats. The maximum concentration of vinorelbine used, 5 mg/mL with 0.1 mL, equates to a dose of 1.56 mg/kg in rats, much less than the toxicity dosage.

As mentioned in the background, animal models of chemical phlebitis have primarily been established by intravascular injection vesicants via the marginal ear vein in rabbits or the tail vein in mice. Additionally, rats have been used by injection vinorelbine gel around the tail vein21, or intradermal injection docetaxel and vinorelbine in lower extremities22 to induce chemical phlebitis. The rat’s dorsal pedal vein with large and superficial vessel runs along the dorsal paw surface, originating from the metatarsal veins and ascending laterally to join the saphenous vein in the hindlimb23. So, the distal hindlimb, with superficial venous accessibility and abundant surrounding muscle tissue, provides a more clinically relevant anatomical site for studying phlebitis manifestations and pathogenesis. In the present study, intravascular administered a single 0.1 mL of vinorelbine ranging from 1 mg/mL to 5 mg/mL via the dorsal vein of the hind paw in rats displayed erythema, edema and pain as early as day 1 post-injection, with symptom severity escalating in a dose-dependent manner. The edema and pain reached peak level on day 3 accompanied by skin ulcer. While these acute symptoms gradually subsided thereafter but followed with induration and streak formation on day 7. By the end of 7 day, all the rats injected vinorelbine in the hindlimb exhibited manifestation of classical phlebitis at concentrations exceeding 3 mg/mL, as assessed by the phlebitis scale. As a whole, rats administered a single 0.1 mL of vinorelbine with a concentration of 3–5 mg/mL via the dorsal vein of the hind paw can induce redness, swelling, pain, cord-like vein and or skin ulceration lasting for at least 7 days.

To date, there has been few studies about the venous pathological features of chemical phlebitis. In fact, it is indeed challenging to obtain lesioned vein specimens from rabbit ears or mouse tails for histological studies, particularly for examining the microstructure of the tunica intima and endothelial cell morphology. Previous study had reported intimal hyperplasia occurred in a porcine thrombosis model induced by electrical injury and partial occlusion of the jugular vein24. The dorsal pedal vein in rats is analogous to the superficial veins of the human hand due to its large size and superficial accessibility, making it suitable for microstructural examination. In the present study, the venous tissues in rat injected saline presented flexible and thin wall with classical layers of tunica adventitia, media and intima. Meanwhile, the intima displayed hobnailed endothelium and the large lumen was free of thrombus. Obviously, progressive dose-dependent venous damage was observed in chemical phlebitis rat model by injection vinorelbine via dorsal pedal vein. These leison veins were characterized by vascular wall thickening with intimal hyperplasia, luminal stenosis and thrombus. Meanwhile, the neointima showed predominant smooth muscle cells with interspersed collagen. Vascular endothelial cells form the inner layer of blood vessels where they have a key role in the development and maintenance of the functional circulatory system and provide paracrine support to surrounding non-vascular cells25. Endothelial injury is an initial trigger for intimal hyperplasia in vessels, which is a target for developing pharmacological approaches to prevent restenosis26. So the microstructure changes in endothelium were observed by electron microscopy27. The tunica intima of normal vessles presents a smooth, continuous surface with polygonal endothelial cells aligned parallel to the vascular longitudinal axis. In contrast, the intimal surface of veins exposed to vinorelbine in chemical phlebitis rats was rough with abundant blood cells, even densely packed spindle-shaped endothelial cells, and significant accumulation of extracellular matrix. These abnormal morphologies are similar to the injuries in human umbilical vein endothelial cells following 1-hour exposure vinorelbine in vitro28.

The symptomatic and pathological features of chemical phlebitis were characterized in a rat model induced by vinorelbine injection via the dorsal pedal vein. Nevertheless, there are still some limitations. First, the 7-day observation period, while sufficient to capture acute to subacute manifestations such as erythema, edema, pain, and induration, does not reflect the more protracted and complex clinical course seen in patients, which can persist for weeks to months. A longer observation period would be necessary to fully elucidate the natural history of the disease, including its resolution or progression to chronic venous impairment. Second, the underlying molecular mechanisms of vinorelbine-induced phlebitis remain unelucidated. The development of chemical phlebitis entails a complex cascade of events, encompassing endothelial activation, cytokine release, coagulation activation, and inflammatory cell infiltration29. Future investigations should, therefore, incorporate analyses of key molecular mediators-such as inflammatory cytokines, adhesion molecules, and pathways implicated in endothelial dysfunction-to delineate the pathogenesis and identify potential therapeutic targets.

Conclusions

This study successfully establishes a novel rat model of chemical phlebitis in a simplicity and good stability method. The hindlimb of rats are suitable for inducing chemical phlebitis by injection vinorelbine with a safety dose via dorsal pedal vein. The symptoms including erythema, pain, edema, palpable cord, and purulent drainage occurred in the hindlimb of rats administered a single 0.1 mL of vinorelbine with 3–5 mg/mL, which may last for at least 7 days. The pathological features in lesion venous tissues were characterized by vascular wall thickening with intimal hyperplasia, luminal stenosis and thrombus. Therefore, future studies should extend the observation period and incorporate molecular investigations to better understand the long-term progression and underlying mechanisms of vinorebitine-induced phlebitis.