Neonatal hyperleukocytosis and regenerative anemia in a septic puppy

Abstract

This paper reports a case of neonatal hyperleukocytosis in a dog due to a bacterial infection. A 3-week-old, mixed-breed dog was brought to a veterinary college referral center with a history of weight loss despite a good appetite. Clinical and laboratory examinations included: physical examination, complete blood (cell) count (CBC), serum biochemistry profile, abdominal ultrasound examination, and cytology of liver and bone marrow aspirates.

The CBC showed hyperleukocytosis of 158.0 × 10⁹/L (RI: 2.1 to 21.2 × 10⁹/L) and hematocrit of 0.19 L/L (RI: 0.21 to 0.34 L/L). The strong leukemoid reaction was comprised of neutrophils, monocytes, and lymphocytes. The dog was diagnosed with Staphylococcus pseudointermedius liver infection based on liver aspirates and culture. Amoxicillin-clavulanic acid was prescribed. A recheck abdominal ultrasound and CBC repeated 4 wk after initial examination were unremarkable. Neonatal hyperleukocytosis is well-described in human medicine but veterinary studies in small animal neonates are scarce.

Key clinical message:

Hyperleukocytosis in adult dogs may be caused by leukemia or leukemoid reactions. Generalized sepsis is a leading cause of leukemoid reactions in adult dogs and cats. In puppies, neoplasia is less likely, and other causes should be investigated. Similar to human neonates, puppies can mount a strong leukemoid reaction during an infection, even if it is not a generalized septic process.

Hyperleukocytosis in human medicine is defined as a white blood cell (WBC) count > 100 × 10⁹/L (1,2). Different white blood cell types can be elevated, and a list of potential causes can be refined based upon the cell line(s) affected. There are several recognized causes of hyperleukocytosis in veterinary medicine (3). Marked leukocytosis is the hallmark of lymphocytic leukemias (4). Another cause is canine leukocyte-adhesion-deficiency (CLAD), but this is rare in small animals and typically diagnosed in Irish setters (5–7). Neutrophilia is most often due to infectious and non-infectious inflammatory conditions but may also be due to a leukemic process (3). Eosinophilia, in small animals, most commonly occurs with parasitism and hypereosinophilic syndrome, although some degree of eosinophilia can be seen in inflammatory reactions (8); it may also be paraneoplastic or due to myeloid leukemia (3).

Leukemoid reaction is a term used for extreme neutrophilic leukocytosis that resembles leukemia but is not neoplastic in origin (3,4,8). Leukemoid reaction is a well-recognized phenomenon in human neonates with the most common cause being a septic process, rather than a neonatal leukemic process (9–14). Most reports of a leukemoid reaction in veterinary medicine are described in adult animals in cases of canine pyometra (15,16), and steroid-responsive dermatitis in an adult cat (17). A recent report of a leukemoid reaction in a 7-week-old puppy was attributed to a Staphylococcus pseudointermedius infection (18).

The aim of this report was to describe a case of a leukemoid reaction in a neonatal puppy, thus adding to the literature on leukemoid reactions in neonatal puppies.

Case description

A 3-week-old, male intact, golden retriever mixed-breed dog was presented to the Companion Animal Hospital at the Ontario Veterinary College Health Sciences Centre, University of Guelph, with a 1-week history of weight loss. There were 12 puppies in the litter. Two of them died at 1 d of age due to unknown causes. At 2 wk of age, the case puppy started to lose weight. He was still bright and alert, and reportedly nursing well. The remaining puppies in the litter were all gaining weight.

On examination at our hospital, the puppy was bright, alert, and had normal vital signs. He weighed 0.685 kg but was not in poor condition despite reported weight loss. The only abnormality on physical examination was cranial abdominal organomegaly. The breeder reported that the dam and sire were healthy, although it is not known if they were examined by a veterinarian before or after breeding.

Diagnostic tests performed included a complete blood (cell) count (CBC), serum biochemistry profile, abdominal ultrasound examination, fine-needle aspiration of the liver for cytology and culture, a nasal swab for adenovirus polymerase chain reaction (PCR) and a bone marrow aspiration for cytology.

Hematology revealed a severe hyperleukocytosis with a white blood cell count (WBC) of 158.0 × 10⁹/L; reference interval (RI): 2.1 to 21.2 × 10⁹/L. There was marked neutrophilia with a moderate left shift, marked lymphocytosis and marked monocytosis, as well as a macrocytic, normochromic, and regenerative anemia with a hematocrit of 0.19 L/L (RI: 0.21 to 0.34 L/L) (Table 1). The hyperleukocytosis and anemia were confirmed by examination of a blood smear, and although the mean corpuscular hemoglobin concentration (MCHC) was close to the lower limit of the reference interval, the erythrocytes had increased central pallor on the smear (Figure 1). All laboratory testing was performed at the Animal Health Laboratory, University of Guelph, Guelph, Ontario, and cytology was reviewed by a clinical pathologist from the Department of Pathobiology at the Ontario Veterinary College, University of Guelph. The anemia was accompanied by a reticulocytosis of 153.1 × 10⁹/L (RI: ≤ 80 × 10⁹/L) and polychromasia typical at this age. Despite increased central pallor in combination with poikilocytosis, true iron-deficiency was considered unlikely as indicated by MCHC within the reference interval at this age and considering the lack of microcytosis. Neutrophils appeared healthy, but some nuclei were variable in size, and some were misshapen or irregular; consistent with dysplasia. Myeloblasts and promyelocytes were also seen. A serum biochemistry profile showed no abnormalities. The PCR tests for canine herpesvirus and adenovirus from the nasal swab were negative.

Table 1.

Complete blood (cell) count results at the time of the first presentation at 3 wk of age. The reference intervals are for 3-week-old puppies (32).

Parameter (units)	CBC values	Abnormalities	Reference intervals
WBC count (× 109/L)	158.0	High	2.1 to 21.2
RBC (× 1012/L)	2.3		2.3 to 4.9
Hb (g/L)	55	Low	57 to 117
Hematocrit (L/L)	0.19	Low	0.21 to 0.34
Reticulocyte count (× 109/L)	153	High	≤ 80
MCV (fL)	83	High	61 to 80
MCHC (g/L)	282		282 to 394
Seg neutrophils (× 109/L)	72.7	High	2.9 to 10.6
Band neutrophils (× 109/L)	41.1	High	0.0 to 0.3
Lymphocytes (× 109/L)	23.7	High	0.8 to 5.1
Monocytes (× 109/L)	20.5	High	0.0 to 1.1
Platelets (× 109/L)	341		62 to 712

WBC — White blood cells; RBC — Red blood cells; Hb — Hemoglobin; MCV — Mean corpuscular volume; MCHC — Mean corpuscular hemoglobin concentration.

Figure 1.

Blood smear image showing hyperleukocytosis (black arrow) and erythrocytes with increased central pallor (red arrow). Modified Wright’s stain; bar = 20 μm.

An abdominal ultrasound examination was performed initially by a Board-certified internist, and then by a Board-certified radiologist. An enlarged liver with multiple liver nodules was visualized (Figure 2). Evaluation of a fine-needle aspiration sample of the liver nodules revealed cocci bacteria within neutrophils (Figure 3), and the pathologic diagnosis was septic neutrophilic and macrophagic inflammation. The specimen was submitted for bacterial culture. Bone marrow aspiration from the sternum was performed, but the sample was minimally cellular and non-diagnostic. Liver aspirate culture revealed a growth of Staphylococcus pseudointermedius, susceptible to commonly used antibiotics except for tetracycline.

Figure 2.

Ultrasound image of a liver nodule prior to antibiotic therapy. The star indicates the stomach; the distance between the white circles is a diameter of one of the liver nodules (white arrow) which measured 13.4 mm.

Figure 3.

Image of liver aspirate cytology showing intracellular cocci bacteria (black arrow). Modified Wright’s stain; bar = 20 μm.

The puppy was treated with amoxicillin-clavulanic acid (Zoetis, Kirkland, Quebec), 13.75 mg/kg, PO, q12h for 7 d. He was alert and vigorously drinking milk replacer from a bottle while in the hospital; therefore, he was discharged after 24 h. The puppy started to gain weight after 2 d of therapy. The family veterinarian extended the antibiotics for another 7 d, for a total of 14 d. The puppy returned for an abdominal ultrasound examination and repeat blood tests 2 wk after cessation of treatment. The puppy’s weight at the time of recheck was 3.3 kg, and the physical examination was unremarkable. Ultrasound examination did not show any nodules in the liver (Figure 4). Hematology revealed a normal leukogram and resolved anemia, but with mild hypochromia (Table 2).

Figure 4.

Ultrasound image of the liver 4 wk after initial presentation and 2 wk after cessation of antibiotic therapy showing resolution of liver nodules. The orange arrow indicates free fluid, which is considered normal in puppies when present in scant amount.

Table 2.

Complete blood (cell) count results at recheck (2 wk after antibiotic therapy cessation). Reference intervals are for 7-week-old puppies (32).

Parameter (units)	CBC values	Abnormalities	Reference intervals
WBC count (× 109/L)	12.9		6.1 to 25.0
RBC (× 1012/L)	4.3		4.0 to 5.3
Hb (g/L)	94		82 to 116
Hematocrit (L/L)	0.31		0.24 to 0.36
Reticulocyte count (× 109/L)	Not performed		NA
MCV (fL)	73		60 to 73
MCHC (g/L)	301	Low	305 to 334
Seg neutrophil (× 109/L)	6.5		2.9 to 10.6
Band neutrophils (× 109/L)	0.0		0.0 to 0.3
Lymphocytes (× 109/L)	5.0		0.8 to 5.1
Monocytes (×109/L)	0.9		0.0 to 1.1
Eosinophils (× 109/L)	0.52		0.08 to 1.33
Platelets (× 109/L)	479		75 to 634

WBC — White blood cells; RBC — Red blood cells; Hb — Hemoglobin; MCV — Mean corpuscular volume; MCHC — Mean corpuscular hemoglobin concentration; NA — Not available.

Discussion

This case demonstrates that neonatal puppies can mount a marked leukocytosis in response to an infection. Reference intervals for blood count results were different at the 2 times of examination and related to the age of the puppy at the time of analysis.

Hyperleukocytosis may be seen with myeloproliferative disorders, lymphocytic leukemia, and sepsis. Non-septic inflammation, such as one that can occur in severe immune-mediated hemolytic anemia is another cause (19). However, non-leukemic causes rarely lead to a WBC count > 100 × 10⁹/L (19). Leukocyte-adhesion deficiency (LAD) is a rare genetic disorder that is characterized by a deficiency or reduced expression of CD18 (20). It can be detected by genetic testing for canine leukocyte adhesion deficiency (CLAD) that is breed-specific. Three subtypes are recognized in humans: LAD I is characterized by a lack of leukocyte adhesion, LAD II by a lack of leukocyte ability to roll, and LAD III by a defect in leukocyte activation (21). In this case, CLAD gene testing was not performed as LAD typically affects Irish setters (5–7).

Similar to humans, neonatal hyperleukocytosis in this puppy was due to a leukemoid reaction secondary to an infection. Possible points of entry for neonatal infections in dogs are the umbilicus, skin, gastrointestinal, urinary, and respiratory tracts. No signs of omphalitis were observed during the examination, and per the breeder, there were no problems at the umbilical sites of any of the other puppies. Bacterial translocation from the dam’s milk was considered unlikely as the rest of the litter was healthy. No wounds were identified; however, they could have gone unnoticed and healed prior to presentation. The puppy had just started to show dentition, but no obvious infection or pain was identified on oral examination. In the absence of an identifiable infection, a myeloproliferative disorder may have been suspected, especially in view of neutrophil dysplasia (misshapen, irregular nuclei), and circulating myeloblasts and promyelocytes or liver neoplasia in light of the presence of liver nodules.

Staphylococcus pseudointermedius is a Gram-positive cocci that is a common cause of superficial bacterial folliculitis (SBF) in dogs (22). Methicillin-resistant strains are seen especially with prolonged antibiotic use (22). Acute death of a puppy was reported due to S. pseudointermedius infection, in which post-mortem examination showed liver congestion, pleural effusion, and catarrhal gastroenteritis (23). A non-fatal case was recently reported in a 7-week-old pup, in which a leukemoid reaction was also observed with leukocytosis of 165.4 × 10⁹/L (RI: 8.8 to 22.4 × 10⁹/L) (19). This puppy was systemically ill with marked normocytic, normochromic anemia at 0.11 L/L (RI: 0.27 to 0.37 L/L), as well as thrombocytopenia of 109 × 10⁹/L (RI: 193 to 653 × 10⁹/L). In contrast to the current case, that 7-week-old puppy had prominent toxic changes on the blood smear but no dysplastic changes. The anemia, in that case, was more profound than in our case and required a blood transfusion. The differences in severity of the disease, despite the same causative agent, may be in the distribution of the septic process. It is possible that in the current case, the infection was confined to the liver and in the previously reported case, it was systemic. The other possible explanation is that in this case, S. pseudointermedius was the only cultured agent and in the other case, Clostridium bifermentas and S. pseudointermedius were both cultured from liver nodules, with S. pseudointermedius being a potential secondary pathogen not requiring treatment (18).

To the authors’ knowledge, there are no definitive reports of a transfer of S. pseudointermedius from a pet to a human. One extensive study has been conducted at a tertiary hospital in Sweden, where human patients with S. pseudointermedius and their pets, as well as health workers’ pets, were tested, and no pet tested positive (24). The typical site of entry in a person is through a break in the skin barrier, and most cases are reported in immune-compromised individuals. Studies on direct transmission are difficult to undertake, and zoonotic potential for this bacterium exists but the risk is considered low. There is a growing concern worldwide related to methicillin-resistant strains of S. pseudointermedius (MRSP) in humans. It is often highly multi-drug resistant with a worldwide distribution. It is thus crucial to diagnose and treat animal carriers appropriately. Culture should be performed in all cases to prevent improper antimicrobial use and potentially promoting MRSP infections.

There are unique considerations concerning antimicrobial usage in a neonate. There are no studies or reports of illnesses related to antibiotic use in a puppy of such young age. However, in humans, there are many studies linking certain diseases, including asthma, inflammatory bowel disease, and neonatal pneumonia to early age antimicrobial use (25). Because bacterial flora is still developing in neonates, antimicrobial use at that stage can alter its development and have undesirable consequences. This may be even more important in puppies, which are more altricial compared to normal term human neonates. Therefore, antimicrobials should only be used when necessary and ideally based on culture results. The 14-day course used in this case was arbitrary and requested by the breeder. Initially, 1 wk of antibiotics was elected since there is no proven benefit of a more prolonged antibiotic therapy; it could even be detrimental to the patient (26). Unfortunately, the puppy was not rechecked earlier than 4 wk after the initial presentation. Ideally, an abdominal ultrasound re-examination and a CBC would have been performed after 7 d of antibiotic therapy. If all initial abnormalities had resolved, therapy could potentially have been discontinued at that time. This is in accordance with antimicrobial stewardship and in view of the veterinarians’ role in preventing increasing multi-drug resistant bacteria worldwide (27).

In general, anemia in neonates can be caused by blood loss or dam’s iron deficiency (28). Blood loss was not suspected, and no obvious bleeding was identified on examination (including by ultrasound and with no detection of melena or blood in feces). Nursing neonates have low iron stores and do not receive much iron from a milk-based diet, thus a dam’s low iron can contribute even more to anemia (29). Such anemia is typically non-regenerative, which was not the case in this puppy. Moreover, iron sequestration secondary to an infection or inflammation (not a true deficiency) can also be seen (30). Neonatal dogs have a lower hematocrit than adult dogs (31). However, reported neonatal reference intervals have a lower reference limit starting at 0.21 L/L (31), and this puppy had a hematocrit of 0.19 L/L with moderate regeneration and increased central pallor in erythrocytes on a blood smear with MCHC close to the lower limit of the reference interval for a puppy this age (Table 1). Iron deficiency anemia has been documented in neonatal humans born prematurely, but there was no history of a premature birth in this case (32). In the current case, it was suspected that the anemia was due to the sequestration of iron in macrophages because of an infection and not due to an actual iron deficiency; however, sequestration would typically hamper a regenerative response. This is a known phenomenon when the body “hides” its iron stores from the invading bacteria (33–35). This explanation is plausible, as shown by the resolution of anemia after clearing the infection with no need for iron supplementation. It is interesting, that at recheck time, the MCHC was low and MCV decreased from slightly elevated to high end of normal.

In conclusion, this report describes an example of hyperleukocytosis due to a potentially localized infectious process and not caused by neoplasia or generalized sepsis. Human and animal neonates are capable of a marked inflammatory response, and even extremely elevated WBC counts should not lead to an assumption of poor prognosis.