Can You Start IVF Stimulation with a Cold, Fever or Inflammation?

A cold, sore throat, mild fever, vaginitis or cervicitis can arise just before or during an IVF cycle. There is no single international guideline devoted to whether a common cold permits ovarian stimulation. Relevant evidence comes from pandemic statements, retrieval guidance, fertility-assessment recommendations and observational studies. Stimulation, retrieval anaesthesia, puncture-related infection, sperm and early pregnancy require separate assessment.

01 Which part of the IVF cycle does illness affect?

An IVF/ICSI cycle consists of linked steps with different risks. In a typical antagonist protocol, daily gonadotropins begin on menstrual cycle day 2 or 3 and continue for approximately 8–12 days, with ultrasound and hormone monitoring. A trigger is given when follicles are ready, followed about 36 hours later by ultrasound-guided transvaginal oocyte retrieval, usually with intravenous sedation or anaesthesia. Embryos may then be cultured for 5–6 days before fresh transfer or cryopreservation for later transfer.

The question therefore has five components: whether systemic infection during stimulation changes ovarian response or embryo quality; whether respiratory symptoms increase anaesthetic risk on retrieval day; whether genital or pelvic inflammation increases infection after puncture; whether recent male fever affects the semen available on retrieval day; and whether fever or medication around transfer affects implantation or early pregnancy. Evidence and certainty differ across these questions.

This review considers these five components using ESHRE, ASRM, NICE and ACOG documents, Cochrane reviews and peer-reviewed cohorts. The manuscript's evidence search ended on 15 October 2026.

02 The common cold: a gap in direct evidence

The manuscript's search found no study specifically linking a common cold or upper respiratory infection with ovarian-stimulation outcomes. Current stimulation guidance does not discuss this as a separate category.

The closest human evidence concerns SARS-CoV-2. During the large outbreak in China in late 2022, many patients became infected while undergoing stimulation, allowing retrospective and prospective cohort studies. These provide relatively large datasets on acute respiratory infection during stimulation. COVID-19 and a common cold involve different pathogens, systemic inflammation and fever patterns, so their outcomes cannot be treated as interchangeable.

Society responses during the COVID-19 pandemic

On 2 April 2020, ESHRE advised against starting new assisted-reproduction treatment and suggested freezing all embryos for patients already in treatment. From 23 April 2020, services could resume subject to local rules. Its February 2022 update noted the limited continuing relevance of early triage and testing recommendations and referred clinics to national and regional guidance.[1]

ASRM's sixth 2020 update recommended cancellation if an oocyte donor or gestational carrier tested positive before stimulation, retrieval or transfer, with intended parents informed of that possibility beforehand. In high-prevalence settings, planned freeze-all could avoid starting recipient medication before a donor cycle was cancelled.[2]

03 Infection during stimulation: four cohorts

Tian and colleagues' 2023 multicentre retrospective study in JAMA Network Open included 585 couples at seven centres in four Chinese provinces between 1 October and 31 December 2022; 135 couples had at least one partner infected during stimulation. Median oocyte yield was 10 in both groups (P=0.95), and ICSI oocyte-maturity rates were 79.4% and 79.9%. Differences appeared during embryo development: good-quality embryo rates were 47.3% versus 52.1% (OR 0.83, 95% CI 0.71–0.96), and blastocyst formation 33.3% versus 45.3% (OR 0.61, 95% CI 0.52–0.71). Associations persisted after adjustment for age, BMI, infertility duration and cause. Limitations included retrospective design, absence of symptom-severity and viral-load analysis, and no analysis of semen on retrieval day.[3]

Huang and colleagues' 2025 prospective cohort in Reproductive Biology and Endocrinology included 230 couples treated between April 2023 and April 2024, with time-lapse embryo imaging. Most infections were mild: 93/95 infected women and 50/51 infected men. Among 31 couples with both partners infected, blastocyst formation was 73.0%, compared with 85.4% when neither partner was infected; developmental milestones from five cells to expanded blastocyst occurred later. Infection in only one partner showed no clear effect. Among 190 fresh transfers, live-birth rates did not differ statistically between groups.[5]

Hu and colleagues' 2024 matched study in Frontiers in Endocrinology included 202 couples, with infection limited to asymptomatic or mild disease. Oocyte yield was 9.9±6.8 versus 9.7±6.5 (P=0.391), conventional-IVF fertilization 60.8% versus 62.8%, and clinical pregnancy after fresh transfer 53.3% versus 51.5%. Only 15 infected patients underwent fresh transfer. Among 40 inflammatory factors measured, serum MCP-1, TNF-RI and TNF-RII were elevated, while none of the 40 follicular-fluid factors differed statistically between groups.[4]

A 2023 prospective study by Huang and colleagues examined infection before treatment. Of 451 women, 252 had been infected before stimulation, a mean of 63.4±21.9 days earlier. Oocyte yield was 11.4 versus 11.6 and clinical pregnancy 70.3% versus 73.7%. Stratification by intervals of ≤30, 31–60 and >60 days likewise detected no difference.[6]

Why the studies differ

Possible explanations include infection during stimulation versus before treatment; which partner was infected; embryo-development versus pregnancy endpoints; and sample size, particularly the number of fresh transfers. Tian 2023 did not separate symptom severity, whereas Hu 2024 included only mild or asymptomatic infection.

Oocyte yield did not differ in any of the four studies. If acute infection influences outcomes, these findings point more towards embryo development than ovarian response to medication. The difference when both partners were infected in Huang 2025 also makes male infection relevant. The two studies reporting live birth, Hu 2024 and Huang 2025, detected no between-group differences, but their fresh-transfer samples were insufficient to exclude an effect.

04 Fever: oocytes, sperm and early pregnancy

Female fever and the follicular environment

The manuscript's search found no direct human study of fever's effect on oocytes inside the ovary. Hu 2024 offers related evidence: several serum inflammatory factors rose in mild infection, while 40 follicular-fluid factors showed no between-group differences.[4] Mild systemic inflammation may therefore not produce the same measurable changes within follicles. These findings do not establish what happens with high fever or severe infection.

Male fever can affect semen over the following months

Carlsen and colleagues followed 27 healthy men in Human Reproduction in 2003, collecting monthly semen samples for 16 months alongside daily fever records. Fever during meiosis was associated with a 32.6% reduction in sperm concentration (95% CI −49.9 to −9.2); during spermiogenesis, the reduction was 35.0% (95% CI −50.5 to −14.6). During spermiogenesis, normal morphology decreased by 7.4% and immotile sperm increased by 20.4%. Each additional fever day was associated with further concentration reductions of 7.1% and 8.5%, respectively. Other developmental stages showed no significant effect, and individual responses varied greatly.[7]

Sergerie and colleagues reported one previously fertile man in 2007 who had a temperature of 39–40°C for two days. Samples taken before fever and on days 15, 37, 58, 79 and beyond 180 documented transient changes in semen parameters and sperm DNA integrity.[8]

Sperm development through ejaculation takes roughly 64–74 days. Semen on retrieval day therefore reflects the preceding months; fever may still matter after the temperature has normalized. Tian 2023 identified the absence of retrieval-day semen analysis as a limitation.[3]

Fever around conception and early pregnancy

Waller and colleagues' 2018 analysis of the US National Birth Defects Prevention Study examined maternal cold or influenza from 30 days before conception through 60 days afterwards and 30 categories of noncardiac defects. Febrile illness was associated with eight categories, including anencephaly (adjusted OR 1.52, 95% CI 1.11–2.07), spina bifida (1.39, 1.11–1.76), and cleft lip with or without cleft palate (1.23, 1.05–1.45). Cold or influenza without fever showed no association with any of the 30 categories studied.[9]

The authors considered fever a possible contributor, while noting that it could mark more severe infection. This case–control study depended on maternal recall; some findings might have occurred by chance. It did not establish whether antipyretics change these associations.

For IVF, this evidence relates to transfer and early pregnancy. A frozen transfer can be postponed during fever without undoing stimulation and retrieval that have already been completed.

05 Cold medicines and fever treatment

NSAIDs and ovulation

Follicle rupture involves prostaglandins. Nonsteroidal anti-inflammatory drugs (NSAIDs), including ibuprofen, naproxen and meloxicam, inhibit their synthesis and can interfere with ovulation. In Jesam and colleagues' 2014 randomized trial, 56 women took meloxicam for 18 consecutive days beginning around cycle day 5. Ovulatory dysfunction occurred in 55% of cycles at 15 mg/day and 78% at 30 mg/day, demonstrating a dose-related effect.[10]

In conventional stimulation–retrieval cycles, suppression medication and the trigger control timing, and retrieval precedes ovulation. The relevance of this effect is therefore uncertain. This is pharmacological reasoning: the search found no direct study of NSAIDs and oocyte yield during stimulation. For natural-cycle retrieval, mild stimulation, timed intercourse or insemination, ovulation itself is particularly relevant and short-term NSAID use should be discussed with the treating team.

NSAIDs and miscarriage

Li and colleagues' 2018 prospective Kaiser Permanente cohort included 1,097 pregnant women. NSAID use was associated with miscarriage (adjusted HR 1.59), with a stronger association around conception (1.89). Acetaminophen use showed no corresponding association.[11] The underlying reasons for medication—pain, fever or infection—could also affect miscarriage risk; observational data cannot fully remove this confounding.

Antivirals for influenza in pregnancy

ACOG's 2018 Committee Opinion 753 recommended prompt antiviral treatment for suspected or confirmed influenza in pregnancy, regardless of vaccination or pending test results. Starting within 48 hours is ideal, but later presentation should not automatically preclude treatment.[12] Current ACOG guidance continues to favour prompt empiric treatment, with oseltamivir preferred. These recommendations address pregnancy; they do not specifically resolve every circumstance in the luteal phase before a pregnancy test.

“—” indicates no stimulation-specific study identified. These are general medication considerations. Give the treating team the ingredient list of combination products for review.

06 Respiratory symptoms and anaesthesia on retrieval day

Retrieval is commonly performed under intravenous sedation or anaesthesia, often without intubation and with spontaneous breathing. ESHRE's 2019 oocyte-retrieval guidance found no evidence that one sedation method is universally superior; centres and anaesthetists determine their protocols. It describes preoperative fasting, blood-count assessment for anaesthetic safety and undiagnosed bleeding disorders, and observation for about two hours before discharge once oral intake and urination are possible.[14] Follow the centre's specific instructions, including separate rules for solids and clear fluids.

Airway reactivity can increase during and after respiratory infection, with more laryngospasm, bronchospasm or hypoxaemia around anaesthesia. Most systematic evidence comes from children; adult data are limited.[13] Paediatric risk indicators include temperature above 38°C, productive cough, purulent secretions, wheezing and lower-respiratory involvement. These do not establish adult retrieval-specific thresholds.

After a trigger, retrieval is time-sensitive and cannot simply be moved several weeks. Before triggering, the team can still reassess or cancel the cycle. Mild nasal congestion or sore throat should be disclosed to the anaesthetist; fever, productive cough or wheeze require assessment, ideally before the trigger. The team may alter the anaesthetic plan or cancel if safety requires it. These are general clinical considerations rather than a dedicated retrieval guideline for colds.

07 Genital and pelvic inflammation

How common is infection after retrieval?

The retrieval needle passes through the vaginal wall into the ovary, potentially introducing vaginal microorganisms into the pelvis. Bennett and colleagues' 1993 prospective monitoring study reported pelvic infection after 18 of 2,670 retrievals (0.6%), including nine pelvic abscesses.[15] Ludwig and colleagues' 2006 prospective study of 1,058 retrievals found no pelvic infection and one unexplained fever.[16] Levi-Setti and colleagues reviewed 23,827 retrievals at an Italian centre from 1996–2016: 96 complications required inpatient or outpatient treatment, approximately 0.4%, most commonly intra-abdominal bleeding. Infective complications included ovarian abscess, pelvic peritonitis and sepsis.[17]

Infection is uncommon, but an abscess can require hospitalization or surgery. ESHRE 2019 recommends local protocols for prophylactic antibiotics. It does not favour a particular vaginal preparation over saline because superiority or greater safety has not been established. If an endometrioma is inadvertently punctured, the recommendation is to withdraw immediately, flush the needle and change collection tubes.[14]

Screening before treatment

NICE fertility guidance NG257, replacing CG156, section 1.22 recommends offering sensitive Chlamydia trachomatis screening before uterine instrumentation. Patients with a positive result and their partners should be referred for treatment and contact tracing; prophylactic antibiotics should be considered if screening has not been performed.[19] This uterine-instrumentation recommendation should not be presented as an identical rule for every retrieval. CDC's 2021 STI guidance covers chlamydia, gonorrhoea and pelvic inflammatory disease.[20]

Vaginal dysbiosis

Skafte-Holm and colleagues' 2021 systematic review and meta-analysis included 17 studies and 3,543 IVF patients. Dysbiosis prevalence was 18%. It was associated with early pregnancy loss (RR 1.71, 95% CI 1.29–2.27); in molecular-diagnosis subgroups, clinical pregnancy was lower (RR 0.55, 95% CI 0.32–0.93). No significant live-birth effect was found. GRADE certainty was very low for all outcomes.[18]

Hydrosalpinx

Melo and colleagues' 2020 Cochrane review found higher clinical-pregnancy rates after salpingectomy before IVF compared with no surgery in four trials involving 455 women (RR 2.02, 95% CI 1.44–2.82; moderate-certainty evidence). Those trials did not report live birth for this comparison; complication and pelvic-infection estimates were very imprecise.[21] The main concern is transfer and implantation. In suitable cases, stimulation and freezing can precede treatment of the hydrosalpinx, with transfer later; infection and retrieval safety also require assessment.

Chronic endometritis

ESHRE's 2023 recurrent-implantation-failure recommendations state that assessment for chronic endometritis can be considered, using hysteroscopy, haematoxylin–eosin staining and CD138 immunohistochemistry. Antibiotics can be considered after diagnosis. Reviews conflict over whether successful treatment improves pregnancy or live birth.[22]

Systemic chronic inflammatory disease

Lupus and antiphospholipid syndrome are distinct conditions. EULAR's 2017 women's-health recommendations state that assisted reproduction can be undertaken in stable or inactive disease, with assessment of disease activity and thrombosis risk before stimulation.[23]

08 Freeze-all: separating retrieval and transfer

Stimulation and retrieval involve ovarian response and procedural safety; transfer involves implantation and early pregnancy. Cryopreservation can separate these periods.

This explains historical ESHRE and ASRM pandemic freeze-all options.[1][2] If illness develops late in stimulation, one option, after medical and anaesthetic assessment, is retrieval followed by freezing rather than immediate transfer. Freezing does not itself remove the risks of retrieval during illness.

09 A discussion framework by cycle stage

The following framework combines society statements, procedural guidance and cohort evidence. It is an educational synthesis for discussion with clinicians, rather than a recommendation issued by a society.

Before treatment, postponement may mean waiting for another menstrual cycle; stopping after medication has begun also carries medication and time costs. The trigger is an important final opportunity to reassess the plan. Freezing provides a later scheduling option between retrieval and transfer, while clinical safety remains decisive at every stage.

10 What the evidence currently supports

The evidence includes randomized trials and a Cochrane review for salpingectomy in hydrosalpinx, with moderate certainty for clinical pregnancy; NICE recommendations for chlamydia screening and prophylaxis before uterine instrumentation; large retrieval-complication series; longitudinal and case-report data for male fever; and a large case–control study for fever around conception and birth defects. Acute viral infection during stimulation is addressed mainly by four observational COVID-19 cohorts with inconsistent findings.

Important gaps concern common colds and stimulation outcomes, adult retrieval sedation during respiratory infection, and cold-medicine safety specifically during stimulation. Practice in these areas draws on adjacent evidence and centre experience.

A mild cold without systemic illness may permit treatment after assessment. Fever, lower-respiratory symptoms and acute genital infection require particular review, together with the exact stage of the cycle.

11 Evidence table

Studies [3]–[6] concern Chinese centres during the Omicron period and use different endpoints. [8] is a case report, [9] case–control and [11] an observational cohort; none establishes causation. This is a narrative evidence review, not a registered systematic review. The manuscript search ended on 15 October 2026; key guidance was checked for web publication on 24 October 2026.

Medical information and image disclosure

This article is public medical education and does not replace an individual consultation or prescribe a treatment plan. The treating and anaesthetic teams decide whether to start, continue or postpone a cycle according to the patient's condition.

Quantitative findings come from the references, with designs and samples identified in the evidence table. The section 09 framework is an editorial synthesis, not a society recommendation.

The five original medical infographics are retained in Chinese and localized for English and Russian. The cover and six supplied artistic illustrations were made with AI assistance for conceptual communication; they are not clinical images or evidence of treatment outcomes.